sort list_unlist co_code
gen wanted= sum(co_code!=co_code[_n-1])
gen wanted= sum(co_code!=co_code[_n-1])
-
Login or Register
- Log in with
- You are not logged in. You can browse but not post. Login or Register by clicking 'Login or Register' at the top-right of this page. For more information on Statalist, see the FAQ.
-
I first sort the data starting from observations where the variable list_unlist=0. Then I specify a condition: if the previous value of co_code does not match the current value, add 1 to the running sum of wanted. Otherwise, add 0. As the observation before the first observation is missing, the sum will start at 1. The sorting ensures that the sum changes only when order of firm changes. -
-
Dear Andrew Musau
Revisiting this thread again, I have some doubts. In your post #7, based on the data ex, your code
shows that epu is collinear with fixed effects and hence it is dropped.Code:reghdfe nsales epu totexp if list_unlist==0, a(co_code year) vce(robust)
However, I tried this with the below code by not absorbing the years but putting it as an indep-variable as i.year
Based on the above code (where years are not absorbed) the coefficient of epu can be estimated. Let me reproduce it here again with the same dataex in my post #6Code:reghdfe nsales epu totexp i.year if list_unlist==0, a(co_code ) vce(robust)
Qn.1 Why does absorbing (or non-absorbing) years matter in the above results? Also, what is the correct way to communicate to Stata (absorbing years or non-absorbing years).Code:Model 1 reghdfe nsales epu totexp if list_unlist==0, a (co_code year) vce (robust) // years along with firm fixed effects are absorbed Model 2 reghdfe nsales epu totexp i.year if list_unlist==0, a (co_code) vce (robust) // years are not absorbed
Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1(1) (2) VARIABLES yrs_absorbed yrs_NOT_absorbed epu -27.97 (60.40) totexp 0.989*** 0.989*** (0.0520) (0.0520) 2013.year -85.43 (78.58) 2014.year -107.3** (42.00) 2015.year -98.96 (63.09) 2016.year -117.0 (101.7) 2017.year -44.70 (49.18) 2018.year -53.29 (66.28) 2019o.year - o.epu - Constant 53.75 246.6 (131.0) (372.9) Observations 38 38 R-squared 0.999 0.999
I usually stick to xtreg with two way fixed effects, however, in post #7, I didn't understand what you meant or how to check
Hence I thought of using -reghdfe-.In xtreg, fe, you need to change the base level(s) of the FE indicator(s) and check whether the coefficients of any of your variables change to determine if they are collinear with the fixed effects.
Based on some iterations with sample data (#6), I realized that
Am I correct?Code:Model 1 xtreg nsales epu totexp i.year if list_unlist==0, fe vce (robust) Model 2 reghdfe nsales epu totexp i.year if list_unlist==0, a (co_code) cluster (co_code) are exactly the same but Model 3, due to absorbing years is different reghdfe nsales epu totexp if list_unlist==0, a (co_code year) cluster (co_code)
Comment
-
If reghdfe is telling you that the variable epu is collinear with the time effects, then that is the truth. You cannot identify its coefficient in the presence of time effects. If you include both time dummies and the variable epu, then the collinearity may not be apparent. The way you can tell is to change the base level of the time dummies and note that the coefficient of epu also changes.
Code:reghdfe nsales epu totexp i.year if list_unlist==0, a (co_code) cluster (co_code) reghdfe nsales epu totexp ib2017.year if list_unlist==0, a (co_code) cluster (co_code)Comment
-
Dear Stata Members
I would like to compare the betas of the main variable of interest (epu) in two samples and following the excellent post #2 by Andrew Musau ( https://www.statalist.org/forums/for...18#post1450018), I ran the following regressions
My doubtsCode:1. Full Sample xtreg nfa_tass_w epu /// > wcap_w /// > size_w lever_w nw_ta_w /// > ncfo_ta_w sgrowth_w pbitnet_ta_w fcfvol age /// > i.rd_dum i.divid_dum /// > gdpgrowthannual realinterestrate, /// > fe vce (robust) Fixed-effects (within) regression Number of obs = 83,176 Group variable: id Number of groups = 13,966 R-sq: Obs per group: within = 0.0980 min = 1 between = 0.1178 avg = 6.0 overall = 0.1023 max = 15 F(14,13965) = 137.42 corr(u_i, Xb) = -0.0224 Prob > F = 0.0000 (Std. Err. adjusted for 13,966 clusters in id) ---------------------------------------------------------------------------------- | Robust nfa_tass_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -----------------+---------------------------------------------------------------- epu | -.0061424 .0014532 -4.23 0.000 -.0089908 -.003294 wcap_w | -.129883 .0060629 -21.42 0.000 -.1417671 -.1179989 size_w | -.0100739 .0022158 -4.55 0.000 -.0144172 -.0057306 lever_w | .1586407 .0088661 17.89 0.000 .141262 .1760194 nw_ta_w | .0857368 .010008 8.57 0.000 .0661197 .1053539 ncfo_ta_w | .1175199 .0044078 26.66 0.000 .10888 .1261598 sgrowth_w | .0014055 .0005938 2.37 0.018 .0002417 .0025694 pbitnet_ta_w | -.0685863 .0098594 -6.96 0.000 -.0879121 -.0492605 fcfvol | -.0609987 .0106514 -5.73 0.000 -.0818769 -.0401205 age | -.0911009 .0057229 -15.92 0.000 -.1023185 -.0798833 1.rd_dum | .0095375 .002881 3.31 0.001 .0038903 .0151846 1.divid_dum | -.0019662 .0017189 -1.14 0.253 -.0053356 .0014031 gdpgrowthannual | .0016665 .0002305 7.23 0.000 .0012147 .0021183 realinterestrate | -.0012317 .000185 -6.66 0.000 -.0015943 -.0008692 _cons | .6016653 .0172367 34.91 0.000 .5678792 .6354515 -----------------+---------------------------------------------------------------- sigma_u | .2059355 sigma_e | .0912116 rho | .83599994 (fraction of variance due to u_i) ---------------------------------------------------------------------------------- 2. Public firms . xtreg nfa_tass_w epu /// > wcap_w /// > size_w lever_w nw_ta_w /// > ncfo_ta_w sgrowth_w pbitnet_ta_w fcfvol age /// > i.rd_dum i.divid_dum /// > gdpgrowthannual realinterestrate if inrange(wanted,20691,25179), /// > fe vce (robust) Fixed-effects (within) regression Number of obs = 31,678 Group variable: id Number of groups = 3,702 R-sq: Obs per group: within = 0.0994 min = 1 between = 0.0410 avg = 8.6 overall = 0.0521 max = 15 F(14,3701) = 56.19 corr(u_i, Xb) = -0.1401 Prob > F = 0.0000 (Std. Err. adjusted for 3,702 clusters in id) ---------------------------------------------------------------------------------- | Robust nfa_tass_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -----------------+---------------------------------------------------------------- epu | -.0069392 .0020802 -3.34 0.001 -.0110177 -.0028606 wcap_w | -.1325428 .0093545 -14.17 0.000 -.1508833 -.1142023 size_w | -.0107517 .0033258 -3.23 0.001 -.0172722 -.0042311 lever_w | .1418479 .015116 9.38 0.000 .1122114 .1714843 nw_ta_w | .0608775 .0160633 3.79 0.000 .0293838 .0923713 ncfo_ta_w | .1315846 .007731 17.02 0.000 .1164272 .1467421 sgrowth_w | .0002029 .0008345 0.24 0.808 -.0014333 .0018391 pbitnet_ta_w | -.0868343 .0165037 -5.26 0.000 -.1191916 -.054477 fcfvol | -.0745162 .0185201 -4.02 0.000 -.1108268 -.0382056 age | -.1060108 .0093179 -11.38 0.000 -.1242796 -.087742 1.rd_dum | .0170844 .0045316 3.77 0.000 .0081998 .025969 1.divid_dum | .0016121 .0024249 0.66 0.506 -.0031421 .0063664 gdpgrowthannual | .0017148 .0003114 5.51 0.000 .0011043 .0023252 realinterestrate | -.0013304 .0002598 -5.12 0.000 -.0018397 -.000821 _cons | .6949004 .0281028 24.73 0.000 .6398019 .7499989 -----------------+---------------------------------------------------------------- sigma_u | .18927949 sigma_e | .09098976 rho | .81228941 (fraction of variance due to u_i) ---------------------------------------------------------------------------------- 3. Private firms xtreg nfa_tass_w epu /// > wcap_w /// > size_w lever_w nw_ta_w /// > ncfo_ta_w sgrowth_w pbitnet_ta_w fcfvol age /// > i.rd_dum i.divid_dum /// > gdpgrowthannual realinterestrate if inrange(wanted,1,20690), /// > fe vce (robust) Fixed-effects (within) regression Number of obs = 51,498 Group variable: id Number of groups = 10,865 R-sq: Obs per group: within = 0.0946 min = 1 between = 0.1511 avg = 4.7 overall = 0.1441 max = 15 F(14,10864) = 81.32 corr(u_i, Xb) = 0.0732 Prob > F = 0.0000 (Std. Err. adjusted for 10,865 clusters in id) ---------------------------------------------------------------------------------- | Robust nfa_tass_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -----------------+---------------------------------------------------------------- epu | -.0038526 .0020182 -1.91 0.056 -.0078086 .0001034 wcap_w | -.1289886 .008043 -16.04 0.000 -.1447543 -.1132229 size_w | -.0082741 .0030447 -2.72 0.007 -.0142423 -.002306 lever_w | .1693187 .0108317 15.63 0.000 .1480866 .1905508 nw_ta_w | .1060043 .0125875 8.42 0.000 .0813304 .1306781 ncfo_ta_w | .1071755 .0053306 20.11 0.000 .0967265 .1176246 sgrowth_w | .0023622 .0008051 2.93 0.003 .0007841 .0039403 pbitnet_ta_w | -.0561646 .0122278 -4.59 0.000 -.0801333 -.0321959 fcfvol | -.0436219 .0127874 -3.41 0.001 -.0686874 -.0185564 age | -.0827393 .0073675 -11.23 0.000 -.097181 -.0682976 1.rd_dum | .0021617 .0036503 0.59 0.554 -.0049935 .0093169 1.divid_dum | -.0048677 .0023816 -2.04 0.041 -.0095359 -.0001994 gdpgrowthannual | .0016891 .0003327 5.08 0.000 .0010369 .0023413 realinterestrate | -.0010082 .0002591 -3.89 0.000 -.0015162 -.0005003 _cons | .5277511 .0230285 22.92 0.000 .482611 .5728911 -----------------+---------------------------------------------------------------- sigma_u | .21018445 sigma_e | .08997714 rho | .84512415 (fraction of variance due to u_i) ---------------------------------------------------------------------------------- 4. For comparing the impact of EPU on dependent variable between public and private firms . xtreg nfa_tass_w i.group#(c.epu /// > c.wcap_w /// > c.size_w c.lever_w c.nw_ta_w /// > c.ncfo_ta_w c.sgrowth_w c.pbitnet_ta_w c.fcfvol c.age /// > i.rd_dum i.divid_dum /// > c.gdpgrowthannual c.realinterestrate) , /// > fe vce (robust) note: 1.group#1.divid_dum omitted because of collinearity Fixed-effects (within) regression Number of obs = 83,176 Group variable: id Number of groups = 13,966 R-sq: Obs per group: within = 0.0995 min = 1 between = 0.1211 avg = 6.0 overall = 0.1035 max = 15 F(29,13965) = 67.78 corr(u_i, Xb) = -0.0241 Prob > F = 0.0000 (Std. Err. adjusted for 13,966 clusters in id) ------------------------------------------------------------------------------------------ | Robust nfa_tass_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------------------+---------------------------------------------------------------- group#c.epu | 0 | -.0045214 .0020057 -2.25 0.024 -.0084529 -.0005899 1 | -.0072298 .0020663 -3.50 0.000 -.0112801 -.0031795 | group#c.wcap_w | 0 | -.1329142 .0077816 -17.08 0.000 -.1481673 -.1176612 1 | -.1282711 .0090319 -14.20 0.000 -.1459749 -.1105674 | group#c.size_w | 0 | -.0085349 .0027108 -3.15 0.002 -.0138484 -.0032214 1 | -.0107055 .0026186 -4.09 0.000 -.0158383 -.0055726 | group#c.lever_w | 0 | .1663109 .0105582 15.75 0.000 .1456154 .1870064 1 | .147942 .0145136 10.19 0.000 .1194935 .1763905 | group#c.nw_ta_w | 0 | .1095518 .0120286 9.11 0.000 .0859742 .1331295 1 | .0593189 .0153008 3.88 0.000 .0293272 .0893106 | group#c.ncfo_ta_w | 0 | .1064067 .0053061 20.05 0.000 .096006 .1168074 1 | .1359096 .0077406 17.56 0.000 .120737 .1510822 | group#c.sgrowth_w | 0 | .0023495 .0008009 2.93 0.003 .0007797 .0039194 1 | .0000162 .0008481 0.02 0.985 -.0016462 .0016787 | group#c.pbitnet_ta_w | 0 | -.0570506 .0121774 -4.68 0.000 -.0809199 -.0331813 1 | -.0849207 .0162361 -5.23 0.000 -.1167455 -.0530958 | group#c.fcfvol | 0 | -.0526131 .0127081 -4.14 0.000 -.0775226 -.0277036 1 | -.0745858 .0183586 -4.06 0.000 -.1105712 -.0386004 | group#c.age | 0 | -.0883176 .0063378 -13.94 0.000 -.1007405 -.0758947 1 | -.0994264 .0075975 -13.09 0.000 -.1143186 -.0845342 | group#rd_dum | 0 1 | .0044991 .003565 1.26 0.207 -.0024887 .0114869 1 0 | .0870297 .0328445 2.65 0.008 .0226501 .1514092 1 1 | .1027574 .0334029 3.08 0.002 .0372832 .1682316 | group#divid_dum | 0 1 | -.0049005 .0023522 -2.08 0.037 -.0095111 -.0002899 1 0 | -.0013391 .0023965 -0.56 0.576 -.0060366 .0033584 1 1 | 0 (omitted) | group#c.gdpgrowthannual | 0 | .0016324 .000333 4.90 0.000 .0009797 .0022851 1 | .0016528 .0003112 5.31 0.000 .0010428 .0022628 | group#c.realinterestrate | 0 | -.0010018 .0002597 -3.86 0.000 -.0015109 -.0004927 1 | -.001505 .0002668 -5.64 0.000 -.0020279 -.0009821 | _cons | .5647025 .0220527 25.61 0.000 .5214762 .6079287 -------------------------+---------------------------------------------------------------- sigma_u | .20556076 sigma_e | .09114323 rho | .83570594 (fraction of variance due to u_i) ------------------------------------------------------------------------------------------ test 1.group#c.epu = 0.group#c.epu ( 1) - 0b.group#c.epu + 1.group#c.epu = 0 F( 1, 13965) = 0.89 Prob > F = 0.3448
1. How come the point estimates of model 4 is different from 3 & 2? In the example illustrated in the post 2, this was not the case. Is there anything wrong in my design/codes?
2. Based on model 4, the impact of EPU on investments is more severe for group1 ( t stat is -3.50) than group 0 (t stat is 2.25). However my F test fails to reject that there exists siginificant difference between group 0 and group 1 in terms of EPU on the dependent variable. Does this F statistic (very low figure) makes senses given the difference in t stats of EPU for two groups.
Any help/suggestions in this regard will extremely helpful to me
Comment
-
You have some 0/1 indicators. Enter them as continuous in the interaction.1. How come the point estimates of model 4 is different from 3 & 2? In the example illustrated in the post 2, this was not the case. Is there anything wrong in my design/codes?
Code:i.group#(c.rd_dum c.divid_dum)
After constraining the variances to be equal across the groups, the coefficients do not differ. That does not mean that the variable has no discernible effect on the outcome. To determine that, look at the individual coefficients and their accompanying standard errors.2. Based on model 4, the impact of EPU on investments is more severe for group1 ( t stat is -3.50) than group 0 (t stat is 2.25). However my F test fails to reject that there exists siginificant difference between group 0 and group 1 in terms of EPU on the dependent variable. Does this F statistic (very low figure) makes senses given the difference in t stats of EPU for two groups.Comment
-
Thanks Andrew Musau for the instant help. However, it seems that still I have the same issues while running the commands which give different coefficients for subsample compared to group-wise interaction. For easiness, I dropped indicator variables and run the following commands. I am also not showing complete variables in the results for the purpose of brevity
Code:1. Listed xtreg nfa_tass_w epu /// > wcap_w /// > size_w lever_w nw_ta_w /// > ncfo_ta_w sgrowth_w pbitnet_ta_w fcfvol age /// > gdpgrowthannual realinterestrate if inrange(wanted,20691,25179), /// > fe vce (robust) Fixed-effects (within) regression Number of obs = 31,678 Group variable: id Number of groups = 3,702 R-sq: Obs per group: within = 0.0981 min = 1 between = 0.0350 avg = 8.6 overall = 0.0457 max = 15 F(12,3701) = 64.79 corr(u_i, Xb) = -0.1642 Prob > F = 0.0000 (Std. Err. adjusted for 3,702 clusters in id) ---------------------------------------------------------------------------------- | Robust nfa_tass_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -----------------+---------------------------------------------------------------- epu | -.0068699 .0020765 -3.31 0.001 -.0109411 -.0027986 wcap_w | -.1320722 .0093663 -14.10 0.000 -.1504358 -.1137086 size_w | -.0099582 .0033017 -3.02 0.003 -.0164316 -.0034848 -----------------+---------------------------------------------------------------- sigma_u | .19054013 sigma_e | .09104759 rho | .8141131 (fraction of variance due to u_i) ---------------------------------------------------------------------------------- 2.Unlisted xtreg nfa_tass_w epu /// > wcap_w /// > size_w lever_w nw_ta_w /// > ncfo_ta_w sgrowth_w pbitnet_ta_w fcfvol age /// > gdpgrowthannual realinterestrate if inrange(wanted,1,20690), /// > fe vce (robust) Fixed-effects (within) regression Number of obs = 51,498 Group variable: id Number of groups = 10,865 R-sq: Obs per group: within = 0.0945 min = 1 between = 0.1515 avg = 4.7 overall = 0.1445 max = 15 F(12,10864) = 94.60 corr(u_i, Xb) = 0.0743 Prob > F = 0.0000 (Std. Err. adjusted for 10,865 clusters in id) ---------------------------------------------------------------------------------- | Robust nfa_tass_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -----------------+---------------------------------------------------------------- epu | -.0038325 .0020185 -1.90 0.058 -.0077892 .0001242 wcap_w | -.1293126 .0080474 -16.07 0.000 -.1450871 -.1135382 size_w | -.0082774 .0030462 -2.72 0.007 -.0142484 -.0023063 -----------------+---------------------------------------------------------------- sigma_u | .21015046 sigma_e | .08998242 rho | .84506645 (fraction of variance due to u_i) ---------------------------------------------------------------------------------- 3.Interactions . xtreg nfa_tass_w i.group#(c.epu /// > c.wcap_w /// > c.size_w c.lever_w c.nw_ta_w /// > c.ncfo_ta_w c.sgrowth_w c.pbitnet_ta_w c.fcfvol c.age /// > c.gdpgrowthannual c.realinterestrate) , /// > fe vce (robust) Fixed-effects (within) regression Number of obs = 83,176 Group variable: id Number of groups = 13,966 R-sq: Obs per group: within = 0.0986 min = 1 between = 0.1169 avg = 6.0 overall = 0.0998 max = 15 F(24,13965) = 81.18 corr(u_i, Xb) = -0.0349 Prob > F = 0.0000 (Std. Err. adjusted for 13,966 clusters in id) ------------------------------------------------------------------------------------------ | Robust nfa_tass_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------------------+---------------------------------------------------------------- group#c.epu | 0 | -.0070252 .0019884 -3.53 0.000 -.0109227 -.0031277 1 | -.0048869 .0020196 -2.42 0.016 -.0088457 -.0009282 | group#c.wcap_w | 0 | -.1356046 .0077434 -17.51 0.000 -.1507827 -.1204265 1 | -.1228751 .0087487 -14.04 0.000 -.1400237 -.1057264 | group#c.size_w | 0 | -.0098321 .0026062 -3.77 0.000 -.0149406 -.0047237 1 | -.0091169 .0025275 -3.61 0.000 -.0140711 -.0041627 | -------------------------+---------------------------------------------------------------- sigma_u | .20602196 sigma_e | .09118434 rho | .83619693 (fraction of variance due to u_i) ----------------------------------------------------------------------------------------- ( 1) - 0b.group#c.epu + 1.group#c.epu = 0 F( 1, 13965) = 0.59 Prob > F = 0.4407 -
I am really puzzled why this happens and I suspect in my sample there might be some firms that may not remain continuously listed/unlisted during the sample period and such transition might be driving these results. I don't know how to pick only those firms that remained within that particular status during the sample period.
I have another question too,
I am sorry I couldn't understand this.After constraining the variances to be equal across the groups, the coefficients do not differ. That does not mean that the variable has no discernible effect on the outcome. To determine that, look at the individual coefficients and their accompanying standard errors.
Can I say that
The impact of epu on the dependent variables is more stronger in group 0 vis-a-vis with group 1, however, assuming equal variance, the difference between groups is not significant. Or what is the most appropriate way of describing the above results?Comment
-
Your problem is specific to your data. Present a subsample using dataex that replicates the problem.I am really puzzled why this happens and I suspect in my sample there might be some firms that may not remain continuously listed/unlisted during the sample period and such transition might be driving these results. I don't know how to pick only those firms that remained within that particular status during the sample period.
Clearly, both coefficients are significant in their respective regressions. Whether one has a larger effect than the other, the Wald test says no.I am sorry I couldn't understand this.
Can I say that
The impact of epu on the dependent variables is more stronger in group 0 vis-a-vis with group 1, however, assuming equal variance, the difference between groups is not significant. Or what is the most appropriate way of describing the above results?Comment
-
Dear Prof Andrew Musau
Thank you for your assistance with this matter. I am attaching a sample dataset
clear
Code:input long co_code byte list_unlist int year float(epu casheq_ta_w inven_ta_w tcred_ta_w size_w lever_w) byte(divid_dum rd_dum) 3 0 2004 4.08252 .2525478 . .0152866 5.749393 .0111465 0 0 3 0 2011 4.73533 .5405542 . .0508816 6.389401 . 0 0 3 0 2012 5.20827 .3330311 . .0486552 6.494963 . 0 0 3 0 2013 5.14044 .2445839 . .027342 6.506232 . 0 0 3 0 2014 4.76683 .3417237 . .0426813 6.594413 . 0 0 3 0 2015 4.59869 .327371 . .0321709 6.655569 . 0 0 3 0 2016 4.24885 .0841933 . .0723635 6.677839 .0067959 0 0 3 0 2018 4.13127 .0563302 . .0080312 6.798498 .0131623 0 0 3 0 2019 4.03485 .0432306 . .0101653 6.797047 .0332887 0 0 15 0 2004 4.08252 .0333333 . .2714286 3.044523 . 0 0 289 1 2004 4.08252 .0320513 .0106838 .3600427 6.484941 .3719475 0 0 289 1 2005 4.27638 .0451751 .0090064 .3359542 6.550366 .4125804 0 0 289 1 2006 3.86614 .0246069 .0027665 .3510484 6.532043 .4123471 0 0 289 1 2009 4.95791 .0670172 .0025682 .332518 6.706496 .3393665 0 0 289 1 2010 4.61611 .0462537 .0034218 .3776991 6.742291 .4081416 0 0 289 1 2011 4.73533 .1056295 .0021557 .3394925 7.392401 .5474871 0 0 289 1 2012 5.20827 .0842635 .0041788 .2835308 7.465598 .5553838 0 0 289 1 2013 5.14044 .0311432 .003088 .3884363 7.327781 .513272 1 0 289 1 2014 4.76683 .0305276 .0034084 .4097511 7.207563 .4856995 1 0 289 1 2015 4.59869 .0451696 .005531 .4080936 6.989151 .3811762 0 0 289 1 2016 4.24885 .029749 .0052681 .4293978 6.875335 .3408739 0 0 289 1 2017 4.33206 .0407987 .0082584 .3914705 6.698638 .4061383 0 0 289 1 2018 4.13127 .101127 .0084147 .3543201 6.500539 .3511645 0 0 289 1 2019 4.03485 .1299251 .006889 .2592458 6.718288 .2104182 0 0 328 0 2005 4.27638 .0034722 .6319445 -.0625 3.360375 .2430556 0 0 328 0 2006 3.86614 .0607903 .6504559 -.112462 3.493473 .3617021 0 0 328 0 2007 4.00362 .0314286 .4828571 .2342857 3.555348 .4314286 0 0 328 0 2008 4.27017 .0223124 .3914807 .1825558 3.897924 .4604463 0 0 328 0 2009 4.95791 .026506 .6337349 -.0216867 3.725693 .4313253 0 0 328 0 2010 4.61611 .1296296 .4938271 -.1049383 3.478158 .3055555 0 0 337 0 2004 4.08252 .1580863 .1182318 .4977809 7.026959 .0740281 0 0 337 0 2005 4.27638 .1217654 .0324181 .3804629 7.237922 .0593013 0 0 337 0 2006 3.86614 .2424669 .1342192 .1804577 7.306599 .0548956 0 0 337 0 2007 4.00362 .3100333 .0868144 .1647873 7.352954 .0028191 1 0 337 0 2008 4.27017 .3022244 .0834139 .1543158 7.411194 .0113032 0 0 337 0 2009 4.95791 .2498114 .189288 .1221494 7.451996 .0067893 1 0 337 0 2010 4.61611 .210795 .1739722 .1107405 7.292473 .0085761 1 0 337 0 2011 4.73533 .2007643 .1634366 .1244029 7.289884 .0037532 1 0 337 0 2012 5.20827 .1441376 .1040914 .0910077 7.232299 .0010929 0 0 337 0 2013 5.14044 .1331963 .064063 .1009935 7.285849 .038986 0 0 136909 0 2005 4.27638 . .1677852 . 2.701361 .9194631 0 0 136909 0 2006 3.86614 . .1904762 .0357143 3.226844 .9444444 0 0 136909 0 2010 4.61611 .0192308 .923077 . 2.341806 .8365385 0 0 136909 0 2011 4.73533 .0078125 .9453125 . 2.549445 .859375 0 0 136909 0 2012 5.20827 .2899408 .6804734 . 2.827314 .6686391 0 0 136911 0 2005 4.27638 .0181818 . . 1.902107 . 0 0 136911 0 2006 3.86614 .0064935 . . 2.734367 . 0 0 136911 0 2007 4.00362 . . . 2.901422 . 0 0 136911 0 2008 4.27017 .0055556 .0333333 -.4340241 2.890372 . 0 0 137490 0 2004 4.08252 .0117854 .0728232 .096394 6.343001 .5261214 0 0 137490 0 2005 4.27638 .008985 .1384359 .030782 6.398595 .4767055 0 0 137490 0 2006 3.86614 .0145053 .1788983 -.0024175 6.361475 .4607149 0 0 137490 0 2007 4.00362 .0130124 .2205596 -.094177 6.421297 .4032206 0 0 137490 0 2008 4.27017 .0137183 .2362418 -.0572659 6.440787 .3431169 0 0 137490 0 2011 4.73533 .0203025 .0839172 -.0852707 7.135687 .4697452 0 0 137490 0 2012 5.20827 .022829 .0574763 .0508698 7.21693 .5545768 0 0 137490 0 2013 5.14044 .0452331 .0545117 .0049872 7.452634 .4812689 0 0 137490 0 2014 4.76683 .0223632 .0795613 -.1126761 6.835292 .1595527 0 0 137490 0 2015 4.59869 .0072129 .057365 -.1049251 6.788183 .2066945 0 0 137490 0 2016 4.24885 .0080388 .0333036 -.1369146 6.664026 .2155161 0 0 137490 0 2017 4.33206 .00547 .0866302 -.2054446 6.667084 .2135861 0 0 137490 0 2018 4.13127 .0097522 .0276753 -.1295467 6.631738 .2028202 0 0 137490 0 2019 4.03485 .0046757 .0585062 -.2040523 6.726353 .2126843 0 0 137495 0 2004 4.08252 .0025974 . .1454545 4.343805 .4623376 0 0 137495 0 2005 4.27638 .0027933 . .1522346 4.271095 .5572626 0 0 137495 0 2012 5.20827 .0364026 .1648822 .2633833 3.843744 .0107066 1 0 137495 0 2013 5.14044 .0566728 .1736746 .2687386 4.001864 . 1 0 137495 0 2014 4.76683 .0105448 .1511424 .3198594 4.041296 .0105448 1 0 137495 0 2015 4.59869 .3700787 .1133858 .015748 4.15104 .0062992 1 0 137495 0 2016 4.24885 .3487738 .0967302 .0803815 4.295924 .0027248 0 0 137495 0 2017 4.33206 .3374384 .0862069 .046798 4.396915 . 1 0 137495 0 2018 4.13127 .4354322 .1707577 -.0373533 4.540098 . 1 0 137495 0 2019 4.03485 .4565425 .1671442 .0659026 4.651099 . 1 0 137511 0 2004 4.08252 .0253749 .1280277 .0868897 5.561066 .509804 0 0 137511 0 2005 4.27638 .0297641 .184029 .1150635 5.618587 .4533575 0 0 137511 1 2006 3.86614 .3732809 .1366315 .0430434 6.327758 .2239686 0 0 137511 1 2007 4.00362 .0207645 .0900189 .0646531 6.742409 .4567013 0 0 137511 1 2008 4.27017 .0115067 .109712 -.0093357 7.231143 .561514 0 0 137511 1 2009 4.95791 .0212587 .1085552 .0113078 7.34446 .5434867 0 0 137511 1 2010 4.61611 .0224693 .1100539 -.0135848 7.464281 .5219535 0 0 137511 1 2011 4.73533 .0241471 .0934186 .0454126 7.458878 .5746312 0 0 137511 1 2012 5.20827 .0146461 .0735892 .0490794 7.422254 .7055835 0 0 137511 1 2013 5.14044 .0168799 .0692717 -.0021027 7.445476 .7257754 0 0 137511 1 2014 4.76683 .0112241 .0759383 .0345493 7.4446 .7383375 0 0 137511 1 2015 4.59869 .0119061 .1065854 .0266036 7.470452 .7245073 0 0 137511 1 2016 4.24885 .0127246 .104445 .0433556 7.4598 .7176417 0 0 137511 1 2017 4.33206 .0123777 .134858 .0396086 7.436381 .6459389 0 0 137511 1 2018 4.13127 .011786 .1779805 -.0253853 7.475793 .5715095 0 0 137511 1 2019 4.03485 .0082707 .16698 .0300125 7.375256 .5567043 0 0 137512 0 2004 4.08252 .0310078 .4573644 . 2.557227 . 0 0 137512 0 2005 4.27638 .0337838 .4256757 . 2.694627 .0067568 0 0 137512 0 2006 3.86614 .0807692 .4269231 . 3.258096 .35 0 0 137512 0 2007 4.00362 .084375 .434375 . 3.465736 .4125 0 0 137512 0 2008 4.27017 .042042 .4294294 . 3.505557 .4174174 0 0 137512 0 2009 4.95791 .0233236 .4460641 . 3.535145 .393586 0 0 137512 0 2010 4.61611 .014881 .4702381 . 3.514526 .375 0 0 137512 0 2011 4.73533 .0426829 .4969512 . 3.490428 .3048781 0 0 137512 0 2012 5.20827 .0560748 .5264798 . 3.468856 .2647975 0 0 137554 1 2004 4.08252 .019474 .2094363 .1571967 8.177122 .6529253 0 1 137554 1 2005 4.27638 .014494 .2259364 .1912448 8.131943 .6735756 0 1 137554 1 2006 3.86614 .0201355 .2490515 .191897 8.213382 .6230353 0 1 137554 1 2007 4.00362 .0174285 .3084341 .152089 8.219407 .5664413 0 1 137554 1 2008 4.27017 .0115084 .3440711 .1157413 8.244308 .5322526 0 1 137554 1 2009 4.95791 .0115957 .3054113 .1534268 8.150988 .63961 0 1 137554 1 2010 4.61611 .0091312 .3328857 .166322 8.22242 .6813482 0 1 137554 1 2011 4.73533 .0081061 .3883488 .1483211 8.268296 .6800657 0 1 137564 0 2005 4.27638 .071441 .2223767 .0097341 7.453504 .4957993 0 1 137564 0 2006 3.86614 .0052349 .2058928 .006013 7.947042 .6089063 0 1 137564 0 2007 4.00362 .0023244 .234267 .0677552 8.14375 .5736533 0 1 137564 0 2008 4.27017 .0079166 .2181533 .1828773 8.298265 .6168936 0 0 137564 0 2009 4.95791 .0017811 .2263751 .1780284 8.262069 .7000594 0 1 137564 0 2010 4.61611 .0029549 .2357492 .183071 8.222178 .738731 0 1 137564 0 2011 4.73533 .0011595 .2398541 .1043299 8.170723 .748692 0 1 137564 0 2012 5.20827 .0039178 .2607522 .1397063 8.144912 .8447095 0 1 137564 0 2013 5.14044 .0031313 .2910943 .1394398 8.173265 .8904624 0 1 137632 0 2004 4.08252 . . .2100841 2.476538 .1932773 0 0 137632 0 2005 4.27638 . . .2173913 2.442347 .1652174 0 0 137632 0 2006 3.86614 . . .2232143 2.415914 .1875 0 0 137632 0 2007 4.00362 .0091743 . .2293578 2.388763 .2018349 0 0 137632 0 2008 4.27017 . . .2336448 2.370244 .1214953 0 0 137632 0 2009 4.95791 .0095238 . .2380952 2.351375 .1142857 0 0 137632 0 2010 4.61611 .0377358 . .2358491 2.360854 .0943396 0 0 137632 0 2011 4.73533 .01 . .29 2.302585 .09 0 0 137688 1 2004 4.08252 .0236771 .0669567 .2840152 7.504447 .3241562 0 0 137688 1 2005 4.27638 .064801 .1407961 -.0558368 8.408382 .3630951 0 0 137688 1 2006 3.86614 .2110976 .1554613 .0351111 8.912608 .3379933 0 0 137688 1 2007 4.00362 .0145976 .1800503 .1102484 9.208448 .2477082 0 0 137688 1 2008 4.27017 .0893457 .2139546 -.1521998 9.473312 .23794 0 0 137688 1 2009 4.95791 .1103242 .1362044 -.1699994 9.751676 .2121108 0 0 137688 1 2010 4.61611 .2361807 .1803601 -.2776453 9.622106 .204322 0 0 137688 1 2011 4.73533 .2018211 .2766977 .1462187 9.857344 .5776177 1 0 137688 1 2012 5.20827 .2263127 .1066368 .0128528 9.617397 .2724193 1 0 137688 1 2013 5.14044 .1102415 .1060889 .0832924 9.697618 .3294039 1 0 137688 1 2014 4.76683 .1140883 .0731478 -.0167311 9.666435 .2777426 1 0 137688 1 2015 4.59869 .1030259 .093911 .0731746 9.580648 .3056733 1 0 137688 1 2016 4.24885 .1370512 .1020521 .0710067 9.433052 .2759592 1 0 137688 1 2017 4.33206 .1151797 .0800547 .0933726 9.518244 .3234453 1 0 137688 1 2018 4.13127 .0407301 .2451776 -.1045302 9.744556 .2153671 1 0 137688 1 2019 4.03485 .0314578 .1357509 .0164746 9.67771 .1530465 1 0 endCode:sort list_unlist co_code //sorting on the basis of list_unlist for co_code gen wanted= sum(co_code!=co_code[_n-1]) xtset co_code year gen group= cond(inrange(wanted, 13, 16), 1, 0) xtreg casheq_ta_w epu /// > tcred_ta_w inven_ta_w /// > size_w lever_w /// > i.rd_dum i.divid_dum /// > if inrange(wanted,13,16) , /// > fe vce (robust) note: 1.rd_dum omitted because of collinearity Fixed-effects (within) regression Number of obs = 52 Group variable: co_code Number of groups = 4 R-sq: Obs per group: within = 0.1636 min = 8 between = 0.8109 avg = 13.0 overall = 0.2686 max = 16 F(3,3) = . corr(u_i, Xb) = 0.3543 Prob > F = . (Std. Err. adjusted for 4 clusters in co_code) ------------------------------------------------------------------------------ | Robust casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | .033932 .0243563 1.39 0.258 -.0435805 .1114445 tcred_ta_w | -.120362 .0194953 -6.17 0.009 -.1824049 -.0583191 inven_ta_w | .1798747 .2523492 0.71 0.527 -.6232131 .9829626 size_w | -.0350561 .0462898 -0.76 0.504 -.1823708 .1122586 lever_w | -.2141343 .2045716 -1.05 0.372 -.8651724 .4369037 1.rd_dum | 0 (omitted) 1.divid_dum | .0295571 .0387905 0.76 0.502 -.0938915 .1530057 _cons | .2767882 .3680514 0.75 0.507 -.8945156 1.448092 -------------+---------------------------------------------------------------- sigma_u | .0273194 sigma_e | .06511714 rho | .14967128 (fraction of variance due to u_i) ------------------------------------------------------------------------------Code:xtreg casheq_ta_w epu /// > tcred_ta_w inven_ta_w /// > size_w lever_w /// > i.rd_dum i.divid_dum /// > if inrange(wanted,1,12) , /// > fe vce (robust) Fixed-effects (within) regression Number of obs = 45 Group variable: co_code Number of groups = 6 R-sq: Obs per group: within = 0.4610 min = 2 between = 0.3840 avg = 7.5 overall = 0.1432 max = 14 F(5,5) = . corr(u_i, Xb) = -0.9171 Prob > F = . (Std. Err. adjusted for 6 clusters in co_code) ------------------------------------------------------------------------------ | Robust casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | -.0732682 .0467978 -1.57 0.178 -.1935659 .0470295 tcred_ta_w | -.4202257 .2351037 -1.79 0.134 -1.024579 .1841275 inven_ta_w | -.1735467 .2077867 -0.84 0.442 -.7076794 .3605861 size_w | .0679587 .0667781 1.02 0.356 -.1036997 .2396172 lever_w | .2651931 .1577577 1.68 0.154 -.1403359 .6707221 1.rd_dum | -.0080575 .0350722 -0.23 0.827 -.0982133 .0820983 1.divid_dum | .0068749 .0377037 0.18 0.862 -.0900455 .1037952 _cons | -.0555718 .2916448 -0.19 0.856 -.8052687 .6941252 -------------+---------------------------------------------------------------- sigma_u | .24453733 sigma_e | .05367831 rho | .95403051 (fraction of variance due to u_i) ------------------------------------------------------------------------------
Code:. xtreg casheq_ta_w i.group#(c.epu /// > c.tcred_ta_w c.inven_ta_w /// > c.size_w c.lever_w /// > c.rd_dum c.divid_dum), /// > fe vce (robust) note: 1.group#c.rd_dum omitted because of collinearity Fixed-effects (within) regression Number of obs = 97 Group variable: co_code Number of groups = 9 R-sq: Obs per group: within = 0.2866 min = 4 between = 0.4398 avg = 10.8 overall = 0.0704 max = 16 F(8,8) = . corr(u_i, Xb) = -0.8664 Prob > F = . (Std. Err. adjusted for 9 clusters in co_code) ------------------------------------------------------------------------------------ | Robust casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------------+---------------------------------------------------------------- group#c.epu | 0 | -.071472 .0463292 -1.54 0.161 -.1783072 .0353632 1 | .0376723 .0250073 1.51 0.170 -.0199947 .0953392 | group#c.tcred_ta_w | 0 | -.4437077 .2259444 -1.96 0.085 -.9647365 .077321 1 | -.0492362 .1155007 -0.43 0.681 -.3155813 .217109 | group#c.inven_ta_w | 0 | -.2534783 .1708262 -1.48 0.176 -.6474042 .1404475 1 | .1809025 .2391912 0.76 0.471 -.3706735 .7324784 | group#c.size_w | 0 | .0314085 .0527345 0.60 0.568 -.0901974 .1530143 1 | -.014071 .0298998 -0.47 0.650 -.0830202 .0548781 | group#c.lever_w | 0 | .2962538 .1537658 1.93 0.090 -.0583308 .6508384 1 | -.2425007 .2381043 -1.02 0.338 -.7915701 .3065687 | group#c.rd_dum | 0 | -.0199276 .0299244 -0.67 0.524 -.0889333 .0490781 1 | 0 (omitted) | group#c.divid_dum | 0 | .0110528 .0402867 0.27 0.791 -.0818485 .1039541 1 | .0124014 .0252841 0.49 0.637 -.0459039 .0707066 | _cons | .1370795 .2178799 0.63 0.547 -.3653525 .6395115 -------------------+---------------------------------------------------------------- sigma_u | .17633983 sigma_e | .06045877 rho | .89481559 (fraction of variance due to u_i) ------------------------------------------------------------------------------------
As an aside, co_code "137511" is 0 during 2004,2005 & 1 during 2007-2016. There are many such companies in my original dataset.
Can I ask one more more question?In my example, two groups are Public (Listed) and Private (Unlisted). So how do I put it? Can I say that impact of epu on dependent variable is significant across both groups (based on post#22), however, there is no significant difference between public and private firms in terms of the effect! Am I correct?Clearly, both coefficients are significant in their respective regressions. Whether one has a larger effect than the other, the Wald test says no.Comment
-
Thanks for the data example. An assumption that underlies this procedure is that you have disjoint groups.
As this is not the case with your example, you have to include an interaction between your panel identifier and the group variable to net out the effect of the non-disjoint groups and thus get the same coefficients.Code:. tab co_code group | group co_code | 0 1 | Total -----------+----------------------+---------- 3 | 9 0 | 9 15 | 1 0 | 1 289 | 0 14 | 14 328 | 6 0 | 6 337 | 10 0 | 10 136909 | 5 0 | 5 136911 | 4 0 | 4 137490 | 14 0 | 14 137495 | 10 0 | 10 137511 | 2 14 | 16 137512 | 9 0 | 9 137554 | 0 8 | 8 137564 | 9 0 | 9 137632 | 8 0 | 8 137688 | 0 16 | 16 -----------+----------------------+---------- Total | 87 52 | 139
Code:xtreg casheq_ta_w epu tcred_ta_w inven_ta_w size_w lever_w i.rd_dum i.divid_dum if inrange(wanted,1,12) , fe xtreg casheq_ta_w epu tcred_ta_w inven_ta_w size_w lever_w i.rd_dum i.divid_dum if inrange(wanted,13,16) , fe xtreg casheq_ta_w i.group#(c.epu c.tcred_ta_w c.inven_ta_w c.size_w c.lever_w c.rd_dum c.divid_dum c.co_code), fe
Res.:
Code:. xtreg casheq_ta_w epu tcred_ta_w inven_ta_w size_w lever_w i.rd_dum i.divid_dum if inra > nge(wanted,1,12) , fe Fixed-effects (within) regression Number of obs = 45 Group variable: co_code Number of groups = 6 R-sq: Obs per group: within = 0.4610 min = 2 between = 0.3840 avg = 7.5 overall = 0.1432 max = 14 F(7,32) = 3.91 corr(u_i, Xb) = -0.9171 Prob > F = 0.0035 ------------------------------------------------------------------------------ casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | -.0732682 .0234109 -3.13 0.004 -.1209546 -.0255818 tcred_ta_w | -.4202257 .1045732 -4.02 0.000 -.6332344 -.2072171 inven_ta_w | -.1735467 .1745541 -0.99 0.328 -.5291017 .1820084 size_w | .0679587 .0460832 1.47 0.150 -.0259098 .1618273 lever_w | .2651931 .1185477 2.24 0.032 .0237194 .5066668 1.rd_dum | -.0080575 .0613975 -0.13 0.896 -.1331202 .1170052 1.divid_dum | .0068749 .0322904 0.21 0.833 -.0588986 .0726483 _cons | -.0555718 .2758036 -0.20 0.842 -.6173653 .5062218 -------------+---------------------------------------------------------------- sigma_u | .24453733 sigma_e | .05367831 rho | .95403051 (fraction of variance due to u_i) ------------------------------------------------------------------------------ F test that all u_i=0: F(5, 32) = 9.60 Prob > F = 0.0000 . . xtreg casheq_ta_w epu tcred_ta_w inven_ta_w size_w lever_w i.rd_dum i.divid_dum if inra > nge(wanted,13,16) , fe note: 1.rd_dum omitted because of collinearity Fixed-effects (within) regression Number of obs = 52 Group variable: co_code Number of groups = 4 R-sq: Obs per group: within = 0.1636 min = 8 between = 0.8109 avg = 13.0 overall = 0.2686 max = 16 F(6,42) = 1.37 corr(u_i, Xb) = 0.3543 Prob > F = 0.2494 ------------------------------------------------------------------------------ casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | .033932 .0293532 1.16 0.254 -.0253052 .0931691 tcred_ta_w | -.120362 .1739219 -0.69 0.493 -.4713505 .2306265 inven_ta_w | .1798747 .2259386 0.80 0.430 -.2760878 .6358372 size_w | -.0350561 .0379964 -0.92 0.361 -.111736 .0416238 lever_w | -.2141343 .1205371 -1.78 0.083 -.4573881 .0291195 1.rd_dum | 0 (omitted) 1.divid_dum | .0295571 .0397218 0.74 0.461 -.0506048 .109719 _cons | .2767882 .2828789 0.98 0.333 -.2940845 .8476608 -------------+---------------------------------------------------------------- sigma_u | .0273194 sigma_e | .06511714 rho | .14967128 (fraction of variance due to u_i) ------------------------------------------------------------------------------ F test that all u_i=0: F(3, 42) = 0.23 Prob > F = 0.8721 . . xtreg casheq_ta_w i.group#(c.epu c.tcred_ta_w c.inven_ta_w c.size_w c.lever_w c.rd_dum c > .divid_dum c.co_code), fe note: 1.group#c.rd_dum omitted because of collinearity note: 1.group#c.co_code omitted because of collinearity Fixed-effects (within) regression Number of obs = 97 Group variable: co_code Number of groups = 9 R-sq: Obs per group: within = 0.2967 min = 4 between = 0.0238 avg = 10.8 overall = 0.0375 max = 16 F(14,74) = 2.23 corr(u_i, Xb) = -0.8322 Prob > F = 0.0140 ------------------------------------------------------------------------------------ casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------------+---------------------------------------------------------------- group#c.epu | 0 | -.0732682 .0263585 -2.78 0.007 -.1257887 -.0207477 1 | .033932 .0272435 1.25 0.217 -.0203518 .0882158 | group#c.tcred_ta_w | 0 | -.4202257 .1177399 -3.57 0.001 -.6548276 -.1856238 1 | -.120362 .1614214 -0.75 0.458 -.4420011 .2012771 | group#c.inven_ta_w | 0 | -.1735467 .196532 -0.88 0.380 -.5651452 .2180519 1 | .1798747 .2096994 0.86 0.394 -.2379605 .5977099 | group#c.size_w | 0 | .0679587 .0518855 1.31 0.194 -.0354255 .1713429 1 | -.0350561 .0352655 -0.99 0.323 -.1053241 .0352119 | group#c.lever_w | 0 | .2651931 .1334739 1.99 0.051 -.0007594 .5311456 1 | -.2141343 .1118736 -1.91 0.059 -.4370474 .0087787 | group#c.rd_dum | 0 | -.0080575 .069128 -0.12 0.908 -.1457981 .1296831 1 | 0 (omitted) | group#c.divid_dum | 0 | .0068749 .0363561 0.19 0.851 -.0655663 .079316 1 | .0295571 .0368668 0.80 0.425 -.0439017 .1030159 | group#c.co_code | 0 | -2.60e-06 2.53e-06 -1.03 0.308 -7.65e-06 2.45e-06 1 | 0 (omitted) | _cons | .229706 .2202313 1.04 0.300 -.2091145 .6685265 -------------------+---------------------------------------------------------------- sigma_u | .18919137 sigma_e | .06043689 rho | .90740206 (fraction of variance due to u_i) ------------------------------------------------------------------------------------ F test that all u_i=0: F(8, 74) = 4.10 Prob > F = 0.0004
The first part is fine. For the second, just keep it simple and state that "... however, we cannot reject the equality of the coefficient on epu across the two groups".In my example, two groups are Public (Listed) and Private (Unlisted). So how do I put it? Can I say that impact of epu on dependent variable is significant across both groups (based on post#22), however, there is no significant difference between public and private firms in terms of the effect! Am I correct?Comment
-
Dear Professor Andrew Musau
Thanks for the wonderful assistance. The term "disjoint group" was almost erased from memory and hence I had to use a long sentence to describe that there certain firms that are in both groups listed and unlisted. I tried that code in my dataset, however, still that problem persists. This time I tried to closely observe what really happens.
1. I tried to pull non-disjoint groups by tab, but -tab-is not working in my data set
I tried the following codes, this time with the variable of interest onlyCode:tab co_code group too many values
***Estimation 1
Estimation 2Code:xtreg casheq_ta_w epu if inrange(wanted,1,20690), fe vce (robust) Fixed-effects (within) regression Number of obs = 138,748 Group variable: co_code Number of groups = 20,042 R-sq: Obs per group: within = 0.0000 min = 1 between = 0.0001 avg = 6.9 overall = 0.0000 max = 16 F(1,20041) = 1.72 corr(u_i, Xb) = -0.0095 Prob > F = 0.1895 (Std. Err. adjusted for 20,042 clusters in co_code) ------------------------------------------------------------------------------ | Robust casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | .0010818 .0008244 1.31 0.189 -.0005341 .0026976 _cons | .0767093 .0037051 20.70 0.000 .069447 .0839716 -------------+---------------------------------------------------------------- sigma_u | .10983069 sigma_e | .07844014 rho | .6622212 (fraction of variance due to u_i) ------------------------------------------------------------------------------ . xtreg casheq_ta_w epu if list_unlist==0, fe vce (robust) Fixed-effects (within) regression Number of obs = 138,748 Group variable: co_code Number of groups = 20,042 R-sq: Obs per group: within = 0.0000 min = 1 between = 0.0001 avg = 6.9 overall = 0.0000 max = 16 F(1,20041) = 1.72 corr(u_i, Xb) = -0.0095 Prob > F = 0.1895 (Std. Err. adjusted for 20,042 clusters in co_code) ------------------------------------------------------------------------------ | Robust casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | .0010818 .0008244 1.31 0.189 -.0005341 .0026976 _cons | .0767093 .0037051 20.70 0.000 .069447 .0839716 -------------+---------------------------------------------------------------- sigma_u | .10983069 sigma_e | .07844014 rho | .6622212 (fraction of variance due to u_i) ------------------------------------------------------------------------------ . xtreg casheq_ta_w epu if group==0, fe vce (robust) Fixed-effects (within) regression Number of obs = 138,748 Group variable: co_code Number of groups = 20,042 R-sq: Obs per group: within = 0.0000 min = 1 between = 0.0001 avg = 6.9 overall = 0.0000 max = 16 F(1,20041) = 1.72 corr(u_i, Xb) = -0.0095 Prob > F = 0.1895 (Std. Err. adjusted for 20,042 clusters in co_code) ------------------------------------------------------------------------------ | Robust casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | .0010818 .0008244 1.31 0.189 -.0005341 .0026976 _cons | .0767093 .0037051 20.70 0.000 .069447 .0839716 -------------+---------------------------------------------------------------- sigma_u | .10983069 sigma_e | .07844014 rho | .6622212 (fraction of variance due to u_i) ------------------------------------------------------------------------------
Code:xtreg casheq_ta_w epu if inrange(wanted,20691,25179), fe vce (robust) Fixed-effects (within) regression Number of obs = 43,447 Group variable: co_code Number of groups = 4,435 R-sq: Obs per group: within = 0.0004 min = 1 between = 0.0000 avg = 9.8 overall = 0.0000 max = 16 F(1,4434) = 7.36 corr(u_i, Xb) = -0.0099 Prob > F = 0.0067 (Std. Err. adjusted for 4,435 clusters in co_code) ------------------------------------------------------------------------------ | Robust casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | .0037048 .001366 2.71 0.007 .0010266 .0063829 _cons | .0492643 .0061085 8.06 0.000 .0372887 .06124 -------------+---------------------------------------------------------------- sigma_u | .08524947 sigma_e | .07510055 rho | .56303976 (fraction of variance due to u_i) ------------------------------------------------------------------------------ . xtreg casheq_ta_w epu if list_unlist==1, fe vce (robust) Fixed-effects (within) regression Number of obs = 43,447 Group variable: co_code Number of groups = 4,435 R-sq: Obs per group: within = 0.0004 min = 1 between = 0.0000 avg = 9.8 overall = 0.0000 max = 16 F(1,4434) = 7.36 corr(u_i, Xb) = -0.0099 Prob > F = 0.0067 (Std. Err. adjusted for 4,435 clusters in co_code) ------------------------------------------------------------------------------ | Robust casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | .0037048 .001366 2.71 0.007 .0010266 .0063829 _cons | .0492643 .0061085 8.06 0.000 .0372887 .06124 -------------+---------------------------------------------------------------- sigma_u | .08524947 sigma_e | .07510055 rho | .56303976 (fraction of variance due to u_i) ------------------------------------------------------------------------------ . xtreg casheq_ta_w epu if group==1, fe vce (robust) Fixed-effects (within) regression Number of obs = 43,447 Group variable: co_code Number of groups = 4,435 R-sq: Obs per group: within = 0.0004 min = 1 between = 0.0000 avg = 9.8 overall = 0.0000 max = 16 F(1,4434) = 7.36 corr(u_i, Xb) = -0.0099 Prob > F = 0.0067 (Std. Err. adjusted for 4,435 clusters in co_code) ------------------------------------------------------------------------------ | Robust casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | .0037048 .001366 2.71 0.007 .0010266 .0063829 _cons | .0492643 .0061085 8.06 0.000 .0372887 .06124 -------------+---------------------------------------------------------------- sigma_u | .08524947 sigma_e | .07510055 rho | .56303976 (fraction of variance due to u_i) ------------------------------------------------------------------------------
*However
I know you and others in the forum have helped me already a lot and there is nothing wrong in those codes you posted, however, if I could ask, could you please tell me what really goes here? As you suggested, I interacted group with co_code but still, some unknown issues are creating troubles here. I know I have asked a lot with regard to the same problem but as this is an important piece of work I thought to be as clear and accurate as possible.Code:xtreg casheq_ta_w i.group#(c.epu c.co_code) ,fe vce (robust) note: 1.group#c.co_code omitted because of collinearity Fixed-effects (within) regression Number of obs = 182,195 Group variable: co_code Number of groups = 23,364 R-sq: Obs per group: within = 0.0001 min = 1 between = 0.0047 avg = 7.8 overall = 0.0030 max = 16 F(3,23363) = 2.56 corr(u_i, Xb) = -0.0846 Prob > F = 0.0533 (Std. Err. adjusted for 23,364 clusters in co_code) --------------------------------------------------------------------------------- | Robust casheq_ta_w | Coef. Std. Err. t P>|t| [95% Conf. Interval] ----------------+---------------------------------------------------------------- group#c.epu | 0 | .0015101 .0007559 2.00 0.046 .0000286 .0029917 1 | .0024331 .0010371 2.35 0.019 .0004002 .0044659 | group#c.co_code | 0 | 6.02e-09 2.41e-08 0.25 0.803 -4.12e-08 5.32e-08 1 | 0 (omitted) | _cons | .0690409 .0050668 13.63 0.000 .0591096 .0789722 ----------------+---------------------------------------------------------------- sigma_u | .10592567 sigma_e | .07811117 rho | .64776034 (fraction of variance due to u_i) ---------------------------------------------------------------------------------Comment
-
Dear Andrew Musau
I tried to extract a sample dataset from my original data to illustrate my problem
Code:input long co_code byte list_unlist int year float(epu casheq_nta size_w lever_w nfa_tass_w) 3 0 2004 4.08252 .3378782 5.749393 .0111465 .1079618 3 0 2011 4.73533 1.176535 6.389401 . .1892527 3 0 2012 5.20827 .4993203 6.494963 . .1444545 3 0 2013 5.14044 .3237737 6.506232 . .173913 3 0 2014 4.76683 .5191189 6.594413 . .1216142 3 0 2015 4.59869 .4867037 6.655569 . .3622442 3 0 2016 4.24885 .0919335 6.677839 .0067959 .1492575 3 0 2018 4.13127 .0596927 6.798498 .0131623 .0965979 3 0 2019 4.03485 .0451839 6.797047 .0332887 .0740617 11 0 2004 4.08252 .0218056 6.644441 .50527 .6019518 11 0 2005 4.27638 .022245 6.548935 .5570508 .7297065 11 0 2006 3.86614 .0246897 6.610561 .537892 .6624041 11 0 2007 4.00362 .0232474 6.739455 .5012425 .5791031 11 1 2008 4.27017 .0323942 6.885919 .4762878 .5133892 11 1 2009 4.95791 .0429088 7.098706 .4842201 .4252313 11 1 2010 4.61611 .0423193 7.367519 .5324872 .3870052 11 1 2011 4.73533 .0380873 7.592971 .5286261 .3545006 11 1 2012 5.20827 .0315372 7.843064 .5681711 .2904631 11 1 2013 5.14044 .0143518 7.939479 .5295627 .4968459 11 1 2014 4.76683 .0209427 8.07331 .5053154 .4579294 11 1 2015 4.59869 .0533765 8.098582 .4703021 .4340689 11 1 2016 4.24885 .0264103 8.114414 .4645005 .4190527 11 1 2017 4.33206 .0273288 8.205601 .3970613 .4261485 11 1 2018 4.13127 .0192138 8.245043 .339801 .4307769 11 1 2019 4.03485 .0131579 8.293124 .291845 .4369041 15 0 2004 4.08252 .0344828 3.044523 . . 16 0 2004 4.08252 .5438596 2.174752 .1931818 . 16 0 2005 4.27638 .6949152 2.302585 .17 . 16 0 2006 3.86614 .9032257 2.4681 .1440678 .4576271 16 0 2007 4.00362 1.266667 2.61007 .125 .3970588 16 0 2009 4.95791 1.935484 2.901422 .0934066 .2967033 16 0 2010 4.61611 .8495576 3.039749 .0813397 .2583732 16 0 2011 4.73533 .008547 3.161247 .0720339 .2288136 16 0 2012 5.20827 .03125 3.273364 .0643939 .2045455 16 0 2015 4.59869 .9105263 3.591818 .046832 .1487603 286 0 2004 4.08252 . 2.484907 . .0333333 286 0 2005 4.27638 .0173913 2.459589 . .025641 286 0 2006 3.86614 .0173913 2.459589 . .017094 286 0 2007 4.00362 .0263158 2.459589 . .008547 286 0 2008 4.27017 .026087 2.4681 . .0084746 286 0 2009 4.95791 .4390244 2.4681 . 0 286 0 2010 4.61611 .245 3.214868 .2449799 .4819277 286 0 2011 4.73533 .4192547 4.515245 .3391685 .2713348 286 0 2012 5.20827 .5595855 4.503138 .2115172 .3499446 286 0 2013 5.14044 .0901328 4.744062 .2367276 .2872063 289 1 2004 4.08252 .0331126 6.484941 .3719475 .3597375 289 1 2005 4.27638 .0473125 6.550366 .4125804 .3428163 289 1 2006 3.86614 .0252276 6.532043 .4123471 .3705591 289 1 2009 4.95791 .0718312 6.706496 .3393665 .3442583 289 1 2010 4.61611 .0484968 6.742291 .4081416 .359056 289 1 2011 4.73533 .1181048 7.392401 .5474871 .316642 289 1 2012 5.20827 .0920172 7.465598 .5553838 .4263552 289 1 2013 5.14044 .0321443 7.327781 .513272 .4315374 289 1 2014 4.76683 .0314888 7.207563 .4856995 .3901897 289 1 2015 4.59869 .0473064 6.989151 .3811762 .328632 289 1 2016 4.24885 .0306611 6.875335 .3408739 .2756947 289 1 2017 4.33206 .0425341 6.698638 .4061383 .29841 289 1 2018 4.13127 .1125042 6.500539 .3511645 .2061608 289 1 2019 4.03485 .1493263 6.718288 .2104182 .1479333 328 0 2005 4.27638 .0034843 3.360375 .2430556 .0520833 328 0 2006 3.86614 .0647249 3.493473 .3617021 .0729483 328 0 2007 4.00362 .0324484 3.555348 .4314286 .0657143 328 0 2008 4.27017 .0228216 3.897924 .4604463 .040568 328 0 2009 4.95791 .0272277 3.725693 .4313253 .0409639 328 0 2010 4.61611 .1489362 3.478158 .3055555 .0308642 337 0 2004 4.08252 .1877702 7.026959 .0740281 .0157998 337 0 2005 4.27638 .1386479 7.237922 .0593013 .2278609 337 0 2006 3.86614 .3200744 7.306599 .0548956 .2058922 337 0 2007 4.00362 .4493453 7.352954 .0028191 .171899 337 0 2008 4.27017 .4331254 7.411194 .0113032 .1542553 337 0 2009 4.95791 .3329981 7.451996 .0067893 .1231358 337 0 2010 4.61611 .2670979 7.292473 .0085761 .1237408 337 0 2011 4.73533 .2511954 7.289884 .0037532 .0982667 337 0 2012 5.20827 .1684121 7.232299 .0010929 .0798034 337 0 2013 5.14044 .1536637 7.285849 .038986 .0499486 362 0 2004 4.08252 .0169333 6.636865 .4025174 .0047201 362 0 2005 4.27638 .0040816 7.422254 .5978001 .0040053 362 0 2006 3.86614 .0243798 7.560184 .6525362 .7592438 362 0 2007 4.00362 .0466235 7.724358 .6266572 .6738996 362 0 2008 4.27017 .0303595 7.749926 .5404928 .8063238 362 0 2009 4.95791 .0119062 7.823526 .5194702 .6925201 362 0 2010 4.61611 .0125134 7.979681 .526532 .5429305 362 0 2011 4.73533 .4916805 7.967869 .4191984 .495791 362 0 2012 5.20827 .0046978 7.716728 . .5730317 362 0 2013 5.14044 .17792 8.07782 .09928 .4746447 362 0 2014 4.76683 .014793 8.164026 . .399795 362 0 2015 4.59869 .0530478 8.397215 . .4793785 362 0 2016 4.24885 .0936963 8.6191 . .4743335 362 0 2017 4.33206 .1102397 8.947325 .1110013 .4774368 362 0 2018 4.13127 .0288534 8.866271 .0159689 .5175347 362 0 2019 4.03485 .0552965 8.846295 . .4873787 363 1 2004 4.08252 .0882985 7.58807 .3197772 .5817675 363 1 2005 4.27638 .131709 7.597196 .263525 .4352605 363 1 2006 3.86614 .1082267 7.929883 .3287755 .4723112 363 1 2007 4.00362 .1233034 8.410499 .4702714 .4363596 363 1 2008 4.27017 .0281212 8.753987 .5174009 .5457315 363 1 2009 4.95791 .0176117 8.974441 .6207809 .6838807 363 1 2010 4.61611 .0252333 8.801936 .5273578 .6747605 363 1 2011 4.73533 .0165435 8.708227 .5200661 .6531681 363 1 2012 5.20827 .0163714 8.670687 .4561111 .6020928 363 1 2013 5.14044 .0128949 8.637622 .4428979 .5483962 363 1 2014 4.76683 .0070748 8.569179 .4632794 .5241133 363 1 2015 4.59869 .0034416 8.55445 .4325215 .4848357 363 1 2016 4.24885 .005138 8.503054 .4135259 .4258997 363 1 2017 4.33206 .0087641 8.414495 .4675975 .392664 363 1 2018 4.13127 .0094206 8.271548 .4139606 .324495 363 1 2019 4.03485 .0076869 8.229298 .4578044 .2986237 365 0 2004 4.08252 .1630571 7.934692 .1146643 .3387645 365 1 2005 4.27638 .1558267 8.191269 .2023107 .2650375 365 1 2006 3.86614 .651152 9.236982 .0970789 .101334 365 1 2007 4.00362 .0936128 9.702588 .2503499 .0799944 365 1 2008 4.27017 .0410973 9.997933 .2375688 .0905919 365 1 2009 4.95791 .0115072 10.64667 .4204245 .1216812 365 1 2010 4.61611 .0051916 10.89567 .5371369 .1111892 365 1 2011 4.73533 .0924486 11.1744 .3144097 .1203663 365 1 2012 5.20827 .0424804 11.1744 .389842 .1127457 365 1 2013 5.14044 .0023936 11.1744 .3722537 .0943177 365 1 2014 4.76683 .0148733 11.1744 .432261 .0715218 365 1 2015 4.59869 .0041329 11.1744 .5916784 .0614459 372 0 2004 4.08252 .0530577 6.739928 .1507984 .7174453 372 0 2005 4.27638 .0563666 6.723953 .0171854 .7499099 372 0 2006 3.86614 .2666408 6.887553 . .5884694 372 0 2009 4.95791 .0492489 7.1142 . .6082323 372 0 2010 4.61611 .1549592 7.183112 . .5577828 372 0 2011 4.73533 .1327897 7.349874 . .5529563 372 0 2012 5.20827 .0224485 7.76612 .4420379 .5383376 372 0 2013 5.14044 .0496535 7.811609 .483091 .8481632 372 0 2014 4.76683 .0378427 7.783099 .4126349 .8526899 372 0 2015 4.59869 .0184691 7.735652 .3805113 .8521302 372 0 2016 4.24885 .0128458 7.678049 .3439641 .8551657 372 0 2017 4.33206 .0142031 7.659549 .2821105 .8344964 372 0 2018 4.13127 .0896421 7.34981 .278167 .6874477 372 0 2019 4.03485 .0815941 7.265639 .2154094 .6334335 381 0 2004 4.08252 .0104369 6.347739 .3739496 .4702381 381 0 2005 4.27638 .0092436 6.653121 .5087719 .6693757 381 0 2006 3.86614 .0081489 6.81947 .3476788 .5937738 381 0 2007 4.00362 .0009817 6.927166 .3128678 .5257944 381 0 2008 4.27017 .0010829 7.091576 .1214844 .5069895 381 0 2009 4.95791 .0007789 7.158436 .1260798 .5244766 381 0 2010 4.61611 .0022766 7.250494 . .3619392 381 0 2011 4.73533 .0564733 7.331977 . .3488615 381 0 2012 5.20827 .0488603 7.594784 .025908 .397374 381 0 2013 5.14044 .2188671 7.93641 .1941158 .3198084 381 0 2014 4.76683 .0505312 8.569823 .2848468 .3423779 381 0 2015 4.59869 .0345745 8.630272 .2552245 .2881435 381 0 2016 4.24885 .0119931 8.55601 .169126 .2571038 381 0 2017 4.33206 .0010505 8.582307 .08264 .2817255 381 0 2018 4.13127 .0272128 8.715421 .0477347 .2591614 381 0 2019 4.03485 .0547364 8.883307 .1542174 .2945386 400 1 2004 4.08252 .0018832 3.974058 .1973684 .5150376 400 1 2005 4.27638 . 4.356709 .1935897 .3576923 400 1 2006 3.86614 . 4.979488 .3555709 .1836314 400 1 2007 4.00362 .1008843 5.612763 .369843 .2698065 400 1 2008 4.27017 .005228 5.846728 .4088414 .2190119 400 1 2009 4.95791 .8418839 6.487684 .2526636 .113242 400 1 2010 4.61611 .0030796 6.122493 .3653509 .7695175 400 1 2011 4.73533 .0033267 6.114567 .3748895 .770336 400 1 2012 5.20827 .003337 6.111467 .3791574 .7716186 400 1 2015 4.59869 .0031348 6.104793 .3861607 .7732143 400 1 2016 4.24885 .0031667 6.094698 .373168 .7794814 414 0 2004 4.08252 .1664145 5.03825 . .1705577 414 0 2005 4.27638 .1530698 4.920711 .0846098 .1823486 414 0 2006 3.86614 .0626186 5.123964 .0119048 .1434524 414 0 2007 4.00362 .1157954 5.290789 .0080605 .138539 414 0 2008 4.27017 .27597 5.31763 .0049044 .1172143 414 0 2009 4.95791 .4270353 5.630136 .0014352 .0832436 414 0 2010 4.61611 .2362013 5.718999 . .0768221 414 0 2011 4.73533 .4092542 5.748437 .4351291 .0707683 414 0 2012 5.20827 .4033228 6.194201 .6429884 .0442948 414 0 2013 5.14044 .5807666 6.510705 .724974 .0309386 414 0 2014 4.76683 .5625186 6.265491 .6361909 .0380156 414 0 2015 4.59869 .3456654 6.189085 .586787 .0408289 414 1 2016 4.24885 .2839363 6.094923 .4945897 .0405771 414 1 2017 4.33206 .207671 5.856503 .0071531 .053505 414 1 2018 4.13127 .2719928 5.646859 .0067043 .0144672 414 1 2019 4.03485 .1789925 5.393628 .0036364 .0140909 415 1 2004 4.08252 .2720665 4.808111 .1028571 .1363265 415 1 2005 4.27638 .1927835 4.751 .0319793 .1356958 415 1 2006 3.86614 .119 4.717606 .0768543 .1644325 415 1 2007 4.00362 .1510652 5.183467 .0465508 .1261918 415 1 2008 4.27017 .2024401 5.583872 . .0815483 415 1 2009 4.95791 .9752616 6.02876 . .0491329 415 1 2010 4.61611 1.011036 6.03811 . .0584586 415 1 2011 4.73533 .3191394 6.377747 .0020387 .0411145 415 1 2012 5.20827 .263379 6.539152 . .0346971 415 1 2013 5.14044 .2464338 6.633582 . .318995 415 1 2014 4.76683 .1959306 6.906955 . .2796237 415 1 2015 4.59869 .2192225 7.029088 . .2477414 415 1 2016 4.24885 .6798095 7.029088 . .2399469 415 1 2017 4.33206 .4374342 7.624717 . .2328386 415 1 2018 4.13127 .2095215 7.589133 . .2350501 415 1 2019 4.03485 .6990088 7.532838 . .2446478 425 0 2004 4.08252 . 3.08191 .8348625 .4036697 425 0 2005 4.27638 .0044248 3.122365 .740088 .3348017 425 0 2006 3.86614 .0426829 3.532226 .6812865 .9093567 425 0 2007 4.00362 .042042 3.54674 .5216138 .8645533 425 0 2008 4.27017 .0061538 3.487375 .3211009 .8929664 425 0 2009 4.95791 .0058824 3.532226 .1988304 .8304093 425 0 2010 4.61611 .0028902 3.54674 .0432277 .7982709 425 0 2011 4.73533 .0080429 3.627004 . .7207447 425 0 2012 5.20827 .0070423 3.758872 . .6200466 425 0 2013 5.14044 .0075188 3.693867 . .6492537 425 0 2014 4.76683 .0067416 3.802208 . .5758929 425 0 2015 4.59869 .0101626 3.906005 . .5171026 425 0 2016 4.24885 .0092421 4.000034 . .4652015 425 0 2017 4.33206 .0085034 4.082609 . .4215852 425 0 2018 4.13127 .0060514 4.197202 . .3804511 425 0 2019 4.03485 .0068966 4.29046 . .3465753 520 0 2005 4.27638 .030398 6.488597 .5460767 .1951034 520 0 2006 3.86614 .0194151 6.708206 .4120376 .1593212 520 0 2007 4.00362 .0330303 6.6932 .4123188 .1721403 520 0 2008 4.27017 .033382 6.898311 .4831415 .1840299 520 0 2009 4.95791 .0166217 7.094152 .5549008 .3341356 520 0 2010 4.61611 .0218968 7.289884 .5382831 .3994814 520 0 2011 4.73533 .0156909 7.356599 .5494415 .3839132 520 0 2012 5.20827 .0146034 7.590903 .5308318 .3088733 520 0 2013 5.14044 .0287735 7.560081 .5028646 .3595833 520 0 2014 4.76683 .0350156 7.415296 .5740429 .4169877 520 0 2015 4.59869 .032167 7.511525 .5441771 .372444 520 0 2016 4.24885 .0413258 7.593677 .5707091 .3488114 520 0 2017 4.33206 .0403505 7.592215 .6041762 .3405962 520 0 2018 4.13127 .0335968 7.802127 .6492763 .2673563 520 0 2019 4.03485 .0197133 8.011587 .5898465 .368737 569 0 2004 4.08252 .0153061 4.93663 .2426418 .4486719 569 0 2005 4.27638 .0747229 5.705781 .0525615 .1739854 569 0 2006 3.86614 .0287201 6.279834 .1146712 .075698 569 0 2007 4.00362 .0334978 6.538284 .1199537 .0432644 569 0 2010 4.61611 .0020837 6.358016 .1398371 .0842142 569 0 2011 4.73533 .0020372 5.975081 .2589581 .123507 569 0 2012 5.20827 .0011799 5.827179 .3344137 .1334708 569 0 2013 5.14044 .0018439 5.786897 .3809816 .1202454 569 0 2014 4.76683 .0181181 5.853351 .3220436 .1007463 569 0 2015 4.59869 .0580072 5.943324 .2924974 .0802728 569 0 2016 4.24885 .0197319 5.612763 .2814896 .1014969 569 0 2017 4.33206 .0175289 5.47227 .2932773 .1058824 569 0 2018 4.13127 .0196749 5.47395 .3552852 .1073826 569 0 2019 4.03485 .0240889 5.110782 .513269 .1381182 595 0 2004 4.08252 .0247293 6.551509 .1936313 .2770241 595 0 2005 4.27638 .0823129 6.678971 .0671276 .2615965 595 0 2006 3.86614 .2384549 7.045602 .0485276 .2387175 595 0 2007 4.00362 .1108181 7.378259 .1013868 .1889055 595 0 2008 4.27017 .0233354 7.571834 .0608432 .3046276 595 0 2009 4.95791 .0582805 7.532891 .1256088 .392882 595 0 2010 4.61611 .0762109 7.678789 .0224792 .3130435 595 0 2011 4.73533 .1125698 7.728064 .0180081 .2927527 595 0 2012 5.20827 .1458575 7.995644 .0684636 .2977426 595 0 2013 5.14044 .1082989 8.209009 .2420316 .2272517 595 0 2014 4.76683 .0779349 8.221641 .2923722 .4333978 595 0 2015 4.59869 .0239771 8.268501 .2256162 .3947065 595 0 2016 4.24885 .0985823 8.299485 .1714406 .380004 595 0 2017 4.33206 .0722095 8.27952 .1074753 .3701139 595 0 2018 4.13127 .2092163 8.497357 .065731 .2821413 595 0 2019 4.03485 .2185905 8.829109 .216509 .3014846 612 1 2004 4.08252 .0187266 3.303217 . .0808824 612 1 2005 4.27638 .0035714 3.33577 . .1637011 612 1 2006 3.86614 .003125 3.468856 . .2274143 612 1 2007 4.00362 .002924 3.535145 . .2361516 612 1 2008 4.27017 .0066225 3.819908 . .2192983 612 1 2009 4.95791 .0313253 3.756538 . .3714953 612 1 2010 4.61611 .4977778 3.517498 . .5816024 612 1 2011 4.73533 .0027473 4.695925 .0328767 .3826484 612 1 2012 5.20827 .0020833 4.971894 .1552321 .4234234 612 1 2013 5.14044 .0031586 5.067646 . .4722922 612 1 2014 4.76683 .0017411 5.150977 . .5133256 612 1 2015 4.59869 .0064554 5.144583 . .4816326 612 1 2016 4.24885 .0025349 5.06386 . .4614412 612 1 2017 4.33206 .0019519 5.036952 . .3006493 612 1 2018 4.13127 .0006676 5.009968 . .215477 612 1 2019 4.03485 .0006743 4.999911 . .1597035 641 0 2004 4.08252 .0008052 5.515845 .0575221 . 641 0 2005 4.27638 .0003888 5.550243 .0905558 . 641 0 2006 3.86614 .0003778 5.578975 .0951662 . 641 0 2007 4.00362 .0027111 5.556442 .0969486 . 641 0 2008 4.27017 .0022822 5.574053 .0952562 . 1567 0 2005 4.27638 .0067873 3.795489 .247191 .4134831 1567 0 2006 3.86614 .1381733 3.883624 .2654321 .3641976 1567 0 2009 4.95791 .127193 3.939638 .1848249 .2626459 1567 0 2010 4.61611 .1149425 4.063885 .0309278 .2113402 1567 0 2011 4.73533 .0821918 4.146304 .0443038 .2072785 1567 0 2012 5.20827 .0046368 4.174387 .0292308 .1784616 1567 0 2013 5.14044 .005698 4.25703 .111898 .1968839 1567 0 2014 4.76683 .0178042 4.228292 .0699708 .1661808 1567 0 2015 4.59869 .0184049 4.195697 .1069277 .1415663 1567 0 2016 4.24885 .0218069 4.183576 .0792683 .125 1567 0 2017 4.33206 .0046948 4.162003 .0358255 .1168224 1567 0 2018 4.13127 .0107198 4.189655 .0409091 .1030303 1567 0 2019 4.03485 .0087977 4.231204 .0988372 .1671512 1621 0 2005 4.27638 .0029873 4.900076 .3708116 .4244229 1621 0 2006 3.86614 .015625 4.916325 .3985348 .4871795 1621 0 2007 4.00362 .0010799 5.222516 .5711975 .5323625 1621 0 2008 4.27017 .0027056 5.221976 .5261738 .5779817 1621 0 2009 4.95791 .0022247 5.194067 .4900111 .5593785 1621 0 2010 4.61611 .0017953 5.408068 .5076165 .6066308 1621 0 2011 4.73533 .0007161 5.633002 .4647585 .659034 1621 0 2012 5.20827 . 5.689683 .4060176 .5787694 1621 0 2013 5.14044 .0876847 5.802722 .459541 .4631642 1621 0 2014 4.76683 .0840832 5.856217 .44419 .4547796 1621 1 2015 4.59869 .0819621 5.853638 .4367288 .4292683 1621 1 2016 4.24885 .0394864 5.780126 .310068 .4533663 1621 1 2017 4.33206 .0362059 5.847305 .2821253 .5088074 1621 1 2018 4.13127 .0094659 6.10457 .3016298 .5490065 1621 1 2019 4.03485 .0006588 6.121835 .2912003 .4919903 1803 0 2004 4.08252 .3636364 2.70805 . .0133333 1803 0 2005 4.27638 .0144928 2.639057 . .0857143 1803 0 2006 3.86614 .317757 2.646175 . .0851064 1803 0 2007 4.00362 .4526316 2.624669 . .0869565 1803 0 2010 4.61611 .6464647 2.791165 . .0736196 1803 0 2011 4.73533 .36 2.833213 .0882353 .0705882 1803 0 2012 5.20827 .2554744 2.844909 .0639535 .0697674 1803 0 2013 5.14044 .1360544 2.815409 . .0718563 1803 0 2014 4.76683 .0365854 2.833213 . .0705882 1827 1 2004 4.08252 .0019881 5.018603 .2896825 .2929894 1827 1 2005 4.27638 .0010977 4.513055 .4802632 .3234649 1954 0 2004 4.08252 .0002481 5.999185 .6509551 .8382535 1954 0 2005 4.27638 .0007889 5.941749 .7346295 .8481345 1954 0 2006 3.86614 .0007692 5.966916 .8057904 .7863182 1954 0 2007 4.00362 .0141108 5.942537 .8955107 .7642426 2015 1 2004 4.08252 .0390371 5.073297 .4633688 .2492173 2015 1 2005 4.27638 .0226571 5.696758 .5250084 .3470964 2015 1 2006 3.86614 .1772588 6.827087 .3577236 .1636856 2015 1 2007 4.00362 .0681472 7.294241 .3711781 .1809349 2015 1 2008 4.27017 .2151494 7.913375 .5453248 .2539874 2015 1 2009 4.95791 .0761376 7.892452 .5683601 .2838999 2015 1 2010 4.61611 .0418577 7.814642 .4941452 .4062424 2015 1 2011 4.73533 .034206 7.903227 .4994826 .3946415 2015 1 2012 5.20827 .0504688 8.01099 .4832609 .3111915 2015 1 2013 5.14044 .0311471 7.943534 .4878075 .3599901 2015 1 2014 4.76683 .0309966 8.035927 .4982524 .320356 2015 1 2016 4.24885 .0100354 7.947396 .4971007 .3261439 2216 0 2004 4.08252 .0225828 6.415588 .0300998 .1977752 2216 0 2005 4.27638 .0109548 6.699624 .2186923 .3690433 2216 0 2006 3.86614 .0120892 6.759603 .0927751 .3142758 2216 0 2007 4.00362 .0078472 6.971669 .1184803 .2566604 2216 0 2008 4.27017 .012604 7.114119 .1710055 .3444517 2216 0 2009 4.95791 .0226405 7.025271 .0909576 .3656975 2216 0 2010 4.61611 .070851 7.237419 .0575333 .2636462 2216 0 2011 4.73533 .0097566 7.635255 .0979274 .2029567 2216 0 2012 5.20827 .1302475 7.721437 .0238454 .1662087 2216 0 2013 5.14044 .3100658 7.827759 .0049815 .1411947 2216 0 2014 4.76683 .0016463 8.370988 .2947754 .0777796 2216 0 2015 4.59869 .00143 8.407803 .1905219 .106495 2216 0 2016 4.24885 .0012554 8.696444 .2664003 .0922997 2216 0 2017 4.33206 .001263 8.85026 .2679682 .0847596 2231 0 2004 4.08252 .0126227 5.378052 . .1218837 2231 0 2005 4.27638 .0262009 5.354225 . .1380615 5478 0 2005 4.27638 . 3.218876 .096 .512 5478 0 2006 3.86614 . 3.222868 .0956175 .4820717 5478 0 2007 4.00362 . 3.230804 .1027668 .4624506 5480 0 2005 4.27638 . 3.54674 .3314121 .0057637 5480 0 2006 3.86614 . 3.363842 .2110727 .0069204 5480 0 2007 4.00362 . 3.139833 .012987 .004329 5480 0 2008 4.27017 . 3.122365 .0176211 .0044053 5480 0 2010 4.61611 .1878453 3.7612 .472093 .0023256 5480 0 2012 5.20827 .0014025 4.268298 .710084 0 5480 0 2013 5.14044 .003861 3.258096 .2076923 .0038462 5480 0 2014 4.76683 .007722 3.261935 .2068966 0 5531 1 2004 4.08252 .008 2.533697 . .7936507 5531 0 2005 4.27638 .0247934 2.517696 . .8064516 5560 0 2005 4.27638 .0048077 4.426044 .534689 .4150718 5560 0 2006 3.86614 .0029806 4.20916 .6315007 .46211 5560 0 2009 4.95791 .125 2.379546 .5833333 .7592592 5560 0 2010 4.61611 .0229885 2.186051 .5617978 .8314607 5560 0 2011 4.73533 .05 2.128232 .5 .797619 5560 0 2012 5.20827 .0684932 2.054124 .4615384 .7820513 5560 0 2013 5.14044 .0441176 1.960095 .7183099 .7887324 5568 1 2004 4.08252 .011925 4.777442 .0799663 .6641414 5568 1 2005 4.27638 .0348042 5.114995 .0918919 .4432433 5568 1 2006 3.86614 .0420499 5.066385 .1015132 .4640605 5568 1 2007 4.00362 .2574258 4.844187 .1314961 .4291339 5568 1 2008 4.27017 .3013334 4.986342 .1270492 .3681694 5568 1 2009 4.95791 .3164794 4.945919 .1194879 .3620199 5568 1 2010 4.61611 .4752475 4.898586 .1096197 .3847874 5568 1 2011 4.73533 .4184549 4.884316 .1240545 .3676248 5568 1 2012 5.20827 .4371643 4.896346 .1098655 .3624813 5568 1 2013 5.14044 .2860061 4.835488 .1342335 .3725179 5574 1 2004 4.08252 .022846 5.054333 .4505424 .303127 5574 1 2005 4.27638 .0203725 5.166499 .5185397 .3559612 5574 1 2006 3.86614 .0169014 5.195731 .4880886 .3130194 5574 1 2007 4.00362 .022748 5.415211 .4412811 .2726868 5574 1 2008 4.27017 .0305749 5.736895 .4105127 .3395679 5574 1 2009 4.95791 .0228498 5.987456 .3345884 .3817771 5574 1 2010 4.61611 .0237656 6.095149 .2089249 .3971151 5574 1 2011 4.73533 .0196007 6.331146 .3036668 .3629405 5574 1 2012 5.20827 .0259626 6.364579 .3184713 .3542778 5574 1 2013 5.14044 .025131 6.636996 .3530414 .325118 5574 1 2014 4.76683 .0213954 6.876368 .3799402 .2870705 5574 1 2015 4.59869 .0431369 7.119393 .434086 .2723962 5574 1 2016 4.24885 .1781744 7.416799 .2817215 .2248603 5574 1 2017 4.33206 .1108322 7.596643 .3065529 .3053477 5574 1 2018 4.13127 .0742401 7.730219 .3328061 .3203726 5574 1 2019 4.03485 .0468276 7.831776 .3289275 .3238073 5578 0 2005 4.27638 .0386473 3.7612 .6976744 . 5578 0 2006 3.86614 .005618 4.830711 .9616919 . 5578 0 2008 4.27017 .0092593 3.775057 .6880734 .0344037 5578 0 2009 4.95791 .01417 3.914021 .5988024 .0578842 5578 0 2010 4.61611 .0189873 3.877432 .63147 .0496894 5578 0 2011 4.73533 .0505415 4.757032 .7517182 .0352234 5578 0 2012 5.20827 .0369089 4.498698 .6395996 .054505 5578 0 2013 5.14044 .0013351 4.317488 .76 .06 5632 0 2004 4.08252 .0123457 2.797281 .9634147 . 5632 0 2005 4.27638 .0308219 4.097672 .9053156 . 5632 0 2006 3.86614 .0064103 4.363099 .7821656 . 5632 0 2007 4.00362 .0101523 5.475626 . .0037688 5632 0 2009 4.95791 .0018491 6.700362 .0065215 .9452443 5632 0 2010 4.61611 .002005 6.744766 .0032957 .8525189 5632 0 2011 4.73533 .002764 6.586999 .9017365 .9423925 5632 0 2012 5.20827 .0022714 6.559615 .8681303 .9073654 5632 0 2013 5.14044 .0027439 6.488901 .8400729 .910003 5632 0 2014 4.76683 .0024373 6.549937 .7216819 .8077803 5632 0 2015 4.59869 .0029485 6.522534 .6923416 .7872997 5632 0 2016 4.24885 .0038035 6.451102 .7041364 .8195453 5632 0 2017 4.33206 .0039769 6.319689 .7380266 .9054735 5632 0 2018 4.13127 .0176539 6.424545 .6083009 0 5632 0 2019 4.03485 .0042133 6.429235 .5543004 0 5652 0 2005 4.27638 .122449 1.902107 .2363636 .3272727 5652 0 2006 3.86614 .0363636 1.902107 .1754386 .368421 5652 0 2007 4.00362 .109375 1.960095 .0985916 .2816902 5652 0 2008 4.27017 .0384615 2.091864 .037037 .2592592 5696 0 2004 4.08252 .0178988 4.87367 .5053517 .3379205 5696 0 2005 4.27638 .0159386 5.148076 .333527 .615921 5696 0 2006 3.86614 .0257407 5.352806 .266572 .5804924 5696 0 2007 4.00362 .0052595 5.658436 .2989187 .6675968 5696 0 2008 4.27017 .0119588 5.50289 .2481663 .6454768 5696 0 2009 4.95791 .0044737 6.055847 .3783403 .6887013 5696 0 2010 4.61611 .0061783 6.122493 .3760965 .6217105 5696 0 2011 4.73533 .0303186 5.993961 .311222 .6955112 5696 0 2012 5.20827 .0054602 5.957649 .2937678 .6508921 5696 0 2013 5.14044 .0019943 5.862779 .1970429 .6411715 5696 0 2014 4.76683 .0018577 5.93357 .2280795 .5274172 5696 0 2018 4.13127 .0113223 6.215008 .3168733 .2910836 5696 0 2019 4.03485 .0637021 6.401586 .253152 .228434 5737 0 2004 4.08252 .1166667 1.902107 .1940299 .2089552 5745 0 2004 4.08252 .0182482 3.328627 .749104 .1899642 5745 0 2005 4.27638 .010101 3.401197 .82 .1666667 5745 0 2006 3.86614 .009772 3.433987 .816129 .1516129 5747 1 2004 4.08252 .118274 9.845589 .1915242 .0178067 5747 1 2005 4.27638 .1650763 10.42921 .2480751 .0146925 5747 1 2006 3.86614 .1704987 10.60723 .3265489 .0166303 5747 1 2007 4.00362 .3027779 10.97552 .5176702 .0350391 5747 1 2008 4.27017 .1961892 11.09626 .4483185 .0306338 5747 1 2009 4.95791 .2591647 11.11332 .4781995 .031328 5747 1 2010 4.61611 .2806076 11.1744 .5029912 .0199215 5747 1 2011 4.73533 .0288988 11.1744 .0683945 .0204409 5747 1 2012 5.20827 .0332481 11.1744 .1099389 .0644183 5747 1 2013 5.14044 .1041377 11.1744 .2500194 .0450576 5747 1 2014 4.76683 .0116765 11.1744 .2898492 .0407218 5747 1 2015 4.59869 .0106196 11.1744 .2645762 .0394259 5747 1 2016 4.24885 .0102317 11.1744 .459229 .0738038 5747 1 2017 4.33206 .0280394 11.1744 .483693 .0779245 5747 1 2018 4.13127 .0347027 11.1744 .4263782 .0799709 5747 1 2019 4.03485 .0240252 11.1744 .1916905 .1035798 5759 0 2004 4.08252 .2369386 8.71701 .181775 .1418628 5759 0 2005 4.27638 .3012865 8.908464 .3214098 .2913325 5759 0 2006 3.86614 .203326 9.213943 .2661392 .2782854 5759 0 2007 4.00362 .2271932 9.078407 .290267 .3035942 5759 0 2008 4.27017 .1283518 9.571177 .1367417 .1815089 5759 0 2009 4.95791 .1451413 9.561673 .2862853 .2449964 5759 0 2010 4.61611 .0663218 9.94957 .2815854 .2142318 5759 0 2011 4.73533 .0572502 10.3919 .322762 .1901757 5759 0 2012 5.20827 .2215445 10.60894 .2716241 .1663675 5759 0 2013 5.14044 .3833398 10.97328 .4603761 .2108513 5759 0 2014 4.76683 .2338764 10.87866 .380825 .2595842 5759 0 2015 4.59869 .1254648 10.82355 .3098715 .2739813 5759 0 2016 4.24885 .1073208 11.10097 .3030173 .2768425 5759 0 2017 4.33206 .1369694 11.1744 .3481239 .2589488 5759 0 2018 4.13127 .1233202 11.1744 .2311242 .2333761 5759 0 2019 4.03485 .1154433 11.1744 .1591139 .2568865 5802 1 2004 4.08252 .0426997 4.326778 .3566711 .3606341 5802 1 2005 4.27638 .0804438 4.355426 .3761232 .309371 5802 1 2006 3.86614 .0174292 3.843744 .0149893 .4582441 5802 1 2007 4.00362 .0092166 3.779634 . .4223744 5802 1 2008 4.27017 .0020619 4.576771 .1790123 .4372428 5802 1 2009 4.95791 .0072761 7.328306 .7223535 .6238508 5802 1 2010 4.61611 .010866 7.513546 .6561358 .4967535 5802 1 2011 4.73533 .014049 7.646449 .5940665 .4050736 5802 1 2012 5.20827 .1438759 7.916771 .5290022 .5887564 5802 1 2013 5.14044 .0003106 7.854304 .4962745 .5759857 5802 1 2014 4.76683 .0001835 7.910261 .3871417 .4932663 5802 1 2015 4.59869 .0001845 7.904815 .3892558 .444674 5802 1 2016 4.24885 .0001502 8.110518 .3192863 .327306 5802 1 2017 4.33206 .0059271 8.083822 .3278936 .2640054 5802 1 2018 4.13127 .0000948 8.060066 .3365029 .253546 5802 1 2019 4.03485 .0000953 8.342411 .1528989 .3900005 5876 0 2004 4.08252 .0155451 7.279526 .3417914 .779425 5876 0 2005 4.27638 .0136334 7.231938 .2827392 .8075783 5876 0 2006 3.86614 .015177 7.656195 .4135876 .8592042 5876 0 2007 4.00362 .0156192 7.68928 .3349506 .8221805 5876 0 2008 4.27017 .0212825 7.693026 .2533516 .7911537 5876 0 2009 4.95791 .0205466 7.666596 .2032962 .8374379 5876 0 2010 4.61611 .0189062 7.745868 .187327 .7647059 5876 0 2011 4.73533 .0652576 7.765908 .1365949 .7177378 5876 0 2012 5.20827 .0591817 7.811893 .1252733 .6720787 5876 0 2013 5.14044 .0064448 8.081042 .3034399 .5038978 5876 0 2014 4.76683 .0092191 8.377011 .4670502 .5603774 5876 0 2015 4.59869 .0083278 8.517313 .5758909 .4705035 5876 0 2016 4.24885 .0023554 8.587092 .6421112 .9095674 5876 0 2017 4.33206 .0132874 8.579774 .7332469 .8866407 5876 0 2018 4.13127 .0050254 8.529062 .7923394 .8866336 5876 0 2019 4.03485 .0190441 8.501653 .847732 .8651811 5892 0 2004 4.08252 .2149533 2.564949 .0846154 .1615385 end
Code:sort list_unlist co_code gen wanted= sum(co_code!=co_code[_n-1]) *From 1 to 36 cos are unlisted and 37 to 52 cos are listed gen group= cond(inrange(wanted, 37, 52), 1, 0) xtset co_code year
Code:xtreg casheq_nta epu if inrange(wanted,1,36), fe vce (robust) ------------------------------------------------------------------------------ | Robust casheq_nta | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | .0338644 .0264938 1.28 0.210 -.0199775 .0877063 . . xtreg casheq_nta epu if inrange(wanted,37,52), fe vce (robust) | Robust casheq_nta | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | .0153356 .0340104 0.45 0.659 -.0571558 .0878271 . xtreg casheq_nta i.group#(c.epu c.co_code) ,fe vce (robust) --------------------------------------------------------------------------------- | Robust casheq_nta | Coef. Std. Err. t P>|t| [95% Conf. Interval] ----------------+---------------------------------------------------------------- group#c.epu | 0 | .0308452 .0215167 1.43 0.159 -.0124916 .0741819 1 | .0222761 .0202524 1.10 0.277 -.0185143 .0630665 |Comment
-
Thanks for the additional example. Before, you had only 1 firm that was in both groups, but now you have multiple firms. You want to control for all, so just generate an indicator and interact it with the group variable.
Code:tab co_code group bys co_code (group): gen both = group[1]!= group[_N] replace both= co_code if both xtreg casheq_nta epu if inrange(wanted,1,36), fe xtreg casheq_nta epu if inrange(wanted,37,52), fe xtreg casheq_nta i.group#(c.epu i.both), fe
Res.:
Code:. tab co_code group | group co_code | 0 1 | Total -----------+----------------------+---------- 3 | 9 0 | 9 11 | 4 12 | 16 15 | 1 0 | 1 16 | 9 0 | 9 286 | 10 0 | 10 289 | 0 14 | 14 328 | 6 0 | 6 337 | 10 0 | 10 362 | 16 0 | 16 363 | 0 16 | 16 365 | 1 11 | 12 372 | 14 0 | 14 381 | 16 0 | 16 400 | 0 11 | 11 414 | 12 4 | 16 415 | 0 16 | 16 425 | 16 0 | 16 520 | 15 0 | 15 569 | 14 0 | 14 595 | 16 0 | 16 612 | 0 16 | 16 641 | 5 0 | 5 1567 | 13 0 | 13 1621 | 10 5 | 15 1803 | 9 0 | 9 1827 | 0 2 | 2 1954 | 4 0 | 4 2015 | 0 12 | 12 2216 | 14 0 | 14 2231 | 2 0 | 2 5478 | 3 0 | 3 5480 | 8 0 | 8 5531 | 1 1 | 2 5560 | 7 0 | 7 5568 | 0 10 | 10 5574 | 0 16 | 16 5578 | 8 0 | 8 5632 | 15 0 | 15 5652 | 4 0 | 4 5696 | 13 0 | 13 5737 | 1 0 | 1 5745 | 3 0 | 3 5747 | 0 16 | 16 5759 | 16 0 | 16 5802 | 0 16 | 16 5876 | 16 0 | 16 5892 | 1 0 | 1 -----------+----------------------+---------- Total | 322 178 | 500 . . bys co_code (group): gen both = group[1]!= group[_N] . . xtreg casheq_nta epu if inrange(wanted,1,36), fe Fixed-effects (within) regression Number of obs = 312 Group variable: co_code Number of groups = 35 R-sq: Obs per group: within = 0.0100 min = 1 between = 0.0274 avg = 8.9 overall = 0.0103 max = 16 F(1,276) = 2.80 corr(u_i, Xb) = 0.0427 Prob > F = 0.0957 ------------------------------------------------------------------------------ casheq_nta | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | .0338644 .0202543 1.67 0.096 -.0060082 .073737 _cons | -.0429781 .0907509 -0.47 0.636 -.2216299 .1356737 -------------+---------------------------------------------------------------- sigma_u | .15527318 sigma_e | .14260282 rho | .54245888 (fraction of variance due to u_i) ------------------------------------------------------------------------------ F test that all u_i=0: F(34, 276) = 10.77 Prob > F = 0.0000 . . xtreg casheq_nta epu if inrange(wanted,37,52), fe Fixed-effects (within) regression Number of obs = 176 Group variable: co_code Number of groups = 16 R-sq: Obs per group: within = 0.0021 min = 1 between = 0.0177 avg = 11.0 overall = 0.0009 max = 16 F(1,159) = 0.34 corr(u_i, Xb) = -0.0099 Prob > F = 0.5624 ------------------------------------------------------------------------------ casheq_nta | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- epu | .0153356 .0264194 0.58 0.562 -.0368427 .0675139 _cons | .0339082 .1185502 0.29 0.775 -.200228 .2680443 -------------+---------------------------------------------------------------- sigma_u | .10709629 sigma_e | .13864507 rho | .37369962 (fraction of variance due to u_i) ------------------------------------------------------------------------------ F test that all u_i=0: F(15, 159) = 7.15 Prob > F = 0.0000 . . xtreg casheq_nta i.group#(c.epu i.both), fe note: 1.group#0b.both omitted because of collinearity note: 1.group#11.both omitted because of collinearity note: 1.group#365.both omitted because of collinearity note: 1.group#414.both omitted because of collinearity note: 1.group#1621.both omitted because of collinearity note: 1.group#5531.both omitted because of collinearity Fixed-effects (within) regression Number of obs = 488 Group variable: co_code Number of groups = 46 R-sq: Obs per group: within = 0.0097 min = 1 between = 0.0081 avg = 10.6 overall = 0.0031 max = 16 F(7,435) = 0.61 corr(u_i, Xb) = -0.2190 Prob > F = 0.7494 ------------------------------------------------------------------------------ casheq_nta | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- group#c.epu | 0 | .0338644 .0200507 1.69 0.092 -.0055439 .0732727 1 | .0153356 .0269004 0.57 0.569 -.0375353 .0682065 | group#both | 0 11 | -.0742252 .1687732 -0.44 0.660 -.4059375 .2574871 0 365 | -.0062122 .208941 -0.03 0.976 -.4168715 .4044472 0 414 | -.0137529 .1662131 -0.08 0.934 -.3404335 .3129276 0 1621 | -.0988756 .1660822 -0.60 0.552 -.4252989 .2275477 0 5531 | -.0654156 .2434548 -0.27 0.788 -.5439096 .4130783 1 0 | 0 (omitted) 1 11 | 0 (omitted) 1 365 | 0 (omitted) 1 414 | 0 (omitted) 1 1621 | 0 (omitted) 1 5531 | 0 (omitted) | _cons | -.0123318 .072499 -0.17 0.865 -.1548237 .1301602 -------------+---------------------------------------------------------------- sigma_u | .14747601 sigma_e | .14116906 rho | .52183977 (fraction of variance due to u_i) ------------------------------------------------------------------------------ F test that all u_i=0: F(45, 435) = 9.97 Prob > F = 0.0000 .Last edited by Andrew Musau; 29 Apr 2021, 05:49.Comment
-
Thanks Professor Andrew Musau. The code works perfectly. Thanks a lot. Though I didn't understand the logic of commands, I can readily use them. I may ask for some clarifications based on the above codes later so that I can learn by using them. Thank you once againComment
-
Dear Stata Members
I have doubts concerning the Wald test based on my post #20 -https://www.statalist.org/forums/forum/general-stata-discussion/general/1449783-comparing-coefficients-of-two-equal-fe-regressions-from-two-subsamples?p=1606273#post1606273.
Let me reproduce that Stata output here
Bases on the helpful advice of Andrew Musau , I deduce that the hypothesis (null) for the above results is "Coefficients of EPU are similar across two groups" (Based on post 25). Since F stat is low, we can say that "We cannot reject the equality of coefficients across two groups". I trust my interpretation is correct so farCode:test 1.group#c.epu = 0.group#c.epu ( 1) - 0b.group#c.epu + 1.group#c.epu = 0 F( 1, 13965) = 0.89 Prob > F = 0.3448
Some sources indicate that the Wald test gives Chi-Square stat (https://stats.idre.ucla.edu/other/mu...andor-similar/). My doubt is
1. How different is F-stat from Chi-stat in Wald test? Or specifically what are the assumptions needed for this specific Wald test?Comment
Comment