Is that a way to suppress omitted variables in the regression output?
-
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.
-
-
if you look at the help file for regress you will see that one of the options in "noomitted" -
There are all sorts of cutesy display options that many people may not know about. From within Stata type
help estimation options##display_options-------------------------------------------
Richard Williams
Professor Emeritus of Sociology
University of Notre Dame
StataNow Version: 19.5 MP (2 processor)
EMAIL: [email protected]
WWW: https://academicweb.nd.edu/~rwilliam/Comment
-
As Richard said, there is a lot of ways to do it, I'll just give you one among others, which I find very simple:
If you use the outreg package (ssc install outreg), it automatically deletes the omitted variables.
See the example below :
There is two types of omitted variables : MPG (collinear to mpg), and 0.foreign (base case)Code:sysuse auto.dta,clear gen MPG=mpg reg price mpg MPG weight length i.foreign outreg
They both don't appear in the outreg table output.
Comment
-
Richard, it worked for a linear regression but it's not working for a probit. I checked and it shows as an option on help.
Do you know what could I do?Comment
-
Originally posted by Charlie Joyez View Post
Charlie, I use outreg2 and is not omitting. With noomit option I was able to suppress the omitted variables in the linear regression, but not in the probit.
Do you know if the outreg2 should work as you described?Comment
-
Felipe, can you show your code and output? I have no problem with the following:
Code:. webuse nhanes2f, clear . gen agex = age . probit diabetes i.race age agex, nolog note: agex omitted because of collinearity Probit regression Number of obs = 10,335 LR chi2(3) = 376.66 Prob > chi2 = 0.0000 Log likelihood = -1810.737 Pseudo R2 = 0.0942 ------------------------------------------------------------------------------ diabetes | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- race | Black | .3586926 .0639377 5.61 0.000 .233377 .4840082 Other | .0702952 .1698809 0.41 0.679 -.2626652 .4032557 | age | .0271569 .0016295 16.67 0.000 .0239631 .0303506 agex | 0 (omitted) _cons | -3.168549 .097815 -32.39 0.000 -3.360263 -2.976835 ------------------------------------------------------------------------------ . probit, noomitted note: agex omitted because of collinearity Probit regression Number of obs = 10,335 LR chi2(3) = 376.66 Prob > chi2 = 0.0000 Log likelihood = -1810.737 Pseudo R2 = 0.0942 ------------------------------------------------------------------------------ diabetes | Coef. Std. Err. z P>|z| [95% Conf. Interval] -------------+---------------------------------------------------------------- race | Black | .3586926 .0639377 5.61 0.000 .233377 .4840082 Other | .0702952 .1698809 0.41 0.679 -.2626652 .4032557 | age | .0271569 .0016295 16.67 0.000 .0239631 .0303506 | _cons | -3.168549 .097815 -32.39 0.000 -3.360263 -2.976835 ------------------------------------------------------------------------------ .-------------------------------------------
Richard Williams
Professor Emeritus of Sociology
University of Notre Dame
StataNow Version: 19.5 MP (2 processor)
EMAIL: [email protected]
WWW: https://academicweb.nd.edu/~rwilliam/Comment
-
Also, using outreg after the above probit command,
Make sure both your Stata and outreg are up to date.Code:. outreg ----------------------- diabetes ----------------------- 2bn.race 0.359 (5.61)** 3.race 0.070 (0.41) age 0.027 (16.67)** _cons -3.169 (32.39)** N 10,335 ----------------------- * p<0.05; ** p<0.01-------------------------------------------
Richard Williams
Professor Emeritus of Sociology
University of Notre Dame
StataNow Version: 19.5 MP (2 processor)
EMAIL: [email protected]
WWW: https://academicweb.nd.edu/~rwilliam/Comment
-
Richard,
This is my code:
and this is my outputCode:probit pr_ann_mf_b5 neg_abn_ret roe invest bm i.famafrench, noomit outreg2 using Results/neg_abn_ret.doc, replace drop(i.famafrench i.year) addtext(famafrench FE, YES, Year FE, NO) label sortvar(neg_EPS dec_EPS neg_abn_ret per_neg_EPS_lag EPS_lag bm roe invest)
Code:Probit regression Number of obs = 116824 LR chi2(48) = 11288.79 Prob > chi2 = 0.0000 Log likelihood = -32658.562 Pseudo R2 = 0.1474 ----------------------------------------------------------------------------------------------------------- pr_ann_mf_b5 | Coef. Std. Err. z P>|z| [95% Conf. Interval] ------------------------------------------+---------------------------------------------------------------- neg_abn_ret | .0957623 .0110479 8.67 0.000 .0741088 .1174158 roe | -.0001225 .0004602 -0.27 0.790 -.0010246 .0007796 invest | -.4996485 .0482086 -10.36 0.000 -.5941356 -.4051615 bm | .0000619 .0000845 0.73 0.464 -.0001037 .0002275 | famafrench | Food Products | .375864 .136024 2.76 0.006 .1092619 .642466 Candy & Soda | -.0283262 .1869872 -0.15 0.880 -.3948144 .3381621 Beer & Liquor | 0 (empty) Tobacco Products | 0 (empty) Recreation | .5777728 .1447374 3.99 0.000 .2940926 .8614529 Entertainment | -.0610995 .1453499 -0.42 0.674 -.34598 .2237811 Printing and Publishing | .9284005 .1429658 6.49 0.000 .6481926 1.208608 Consumer Goods | .749213 .1349509 5.55 0.000 .484714 1.013712 Apparel | 1.227236 .1332618 9.21 0.000 .9660475 1.488424 Healthcare | .3732487 .134991 2.76 0.006 .1086713 .6378262 Medical Equipment | .5749243 .1299982 4.42 0.000 .3201326 .829716 Pharmaceutical Products | -.1606609 .1299033 -1.24 0.216 -.4152668 .0939449 Chemicals | .5568309 .1322014 4.21 0.000 .297721 .8159408 Rubber and Plastic Products | .634768 .1448521 4.38 0.000 .3508631 .9186729 Textiles | .1905553 .1822339 1.05 0.296 -.1666166 .5477271 Construction Materials | .2655091 .1365862 1.94 0.052 -.002195 .5332132 Construction | .3477615 .1382199 2.52 0.012 .0768556 .6186675 Steel Works Etc | .6944669 .1359109 5.11 0.000 .4280864 .9608474 Fabricated Products | -.3374416 .2580371 -1.31 0.191 -.8431849 .1683018 Machinery | 1.016925 .1299521 7.83 0.000 .7622239 1.271627 Electrical Equipment | .5314113 .1333215 3.99 0.000 .2701059 .7927167 Automobiles and Trucks | -.1376255 .1449173 -0.95 0.342 -.4216582 .1464071 Aircraft | .1751449 .1602928 1.09 0.275 -.1390233 .489313 Shipbuilding, Railroad Equipment | -.0170481 .2078062 -0.08 0.935 -.4243408 .3902446 Defense | .0628259 .2035048 0.31 0.758 -.3360361 .4616879 Precious Metals | -1.171368 .2158286 -5.43 0.000 -1.594385 -.748352 Non-Metallic and Industrial Metal Mining | -1.237329 .2459552 -5.03 0.000 -1.719393 -.755266 Coal | 0 (empty) Petroleum and Natural Gas | -.3317668 .1326192 -2.50 0.012 -.5916957 -.0718379 Utilities | -1.118873 .3332252 -3.36 0.001 -1.771982 -.4657633 Communication | -.1630523 .1372409 -1.19 0.235 -.4320395 .105935 Personal Services | .8380485 .1346538 6.22 0.000 .5741318 1.101965 Business Services | .9633162 .1278313 7.54 0.000 .7127714 1.213861 Computers | 1.29461 .1291271 10.03 0.000 1.041526 1.547695 Electronic Equipment | 1.207361 .1282248 9.42 0.000 .9560446 1.458676 Measuring and Control Equipment | 1.246006 .1304625 9.55 0.000 .9903043 1.501708 Business Supplies | .4998725 .1398486 3.57 0.000 .2257743 .7739708 Shipping Containers | 1.346132 .1597375 8.43 0.000 1.033052 1.659211 Transportation | .5324629 .1304966 4.08 0.000 .2766943 .7882314 Wholesale | .5552057 .1305404 4.25 0.000 .2993512 .8110603 Retail | 1.19573 .1287813 9.28 0.000 .9433236 1.448137 Restaraunts, Hotels, Motels | .652968 .1337405 4.88 0.000 .3908414 .9150946 Banking | .0550116 .1390205 0.40 0.692 -.2174635 .3274867 Insurance | .2820601 .1339485 2.11 0.035 .0195258 .5445944 Real Estate | -.2168165 .1508831 -1.44 0.151 -.5125419 .0789089 Trading | -.0918848 .1319862 -0.70 0.486 -.3505729 .1668034 Almost Nothing | -.0546181 .1355758 -0.40 0.687 -.3203417 .2111055 | _cons | -1.882865 .1273016 -14.79 0.000 -2.132372 -1.633359 -----------------------------------------------------------------------------------------------------------Comment
-
-
I don't know how to display the outreg table here, but it still shows 3 industries. I believe is the ones empty in this output. In the table they show as omitted.Comment
-
I don't have an example I can use to replicate. But try adding the noemptycells option, e.g.
Code:probit pr_ann_mf_b5 neg_abn_ret roe invest bm i.famafrench, noemptycells noomit
-------------------------------------------
Richard Williams
Professor Emeritus of Sociology
University of Notre Dame
StataNow Version: 19.5 MP (2 processor)
EMAIL: [email protected]
WWW: https://academicweb.nd.edu/~rwilliam/Comment
-
I don't know anything about outreg so can't help with that; however, note that "omitted" and "empty" are different things - try the "noempty" option in addition to the noomitted option
crossed with RW's responseComment
-
It didn't work. Is that any other way to deal with empty cells?Comment
-
see the FAQ for why "it didn't work" is not helpful to readersComment
Comment