Hi, I am recently employing the dynamic spatial panel model to capture some spillover effects, yet I met several questions when using xsmle command in Stata and also the understanding of final outputs.
1. Why do marginal effects get insignificant or inverse signs compared to coefficients?
I regress housing prices on one key coefficient (INS) as well as other controlling variables. The main results show that INS exerts significant positive effects on local housing prices but significant negative spillover effects. However, the marginal results show a different story. Both short-run and long-run indirect effects are not significant. In some cases, even the sign gets changed.
2. I am also confused about the difference between short-run effects and long-term effects, how many years are short-run, and how many are long-run?
I would be grateful if anyone could give me some suggestions.
the code and output are shown below:
1. Why do marginal effects get insignificant or inverse signs compared to coefficients?
I regress housing prices on one key coefficient (INS) as well as other controlling variables. The main results show that INS exerts significant positive effects on local housing prices but significant negative spillover effects. However, the marginal results show a different story. Both short-run and long-run indirect effects are not significant. In some cases, even the sign gets changed.
2. I am also confused about the difference between short-run effects and long-term effects, how many years are short-run, and how many are long-run?
I would be grateful if anyone could give me some suggestions.
the code and output are shown below:
HTML Code:
xsmle ln_hp ins_10k ln_pop ln_GDP_per ln_emp_dens ln_FixedDI ln_indu_str ln_tea_per ln_doc_per ln_gre_per, fe wmat(W_econ_asy) model(sdm) type(ind) nolog dlag(1) effects
HTML Code:
Computing marginal effects standard errors using MC simulation...
Dynamic SDM with spatial fixed-effects Number of obs = 2088
Group variable: city_code Number of groups = 261
Time variable: year Panel length = 8
R-sq: within = 0.4923
between = 0.1081
overall = 0.0894
Mean of fixed-effects = 8.4856
Log-likelihood = .
------------------------------------------------------------------------------
ln_hp | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Main |
ln_hp |
L1. | .6239563 .0174454 35.77 0.000 .5897639 .6581486
|
ins_10k | .0240951 .0026799 8.99 0.000 .0188425 .0293476
ln_pop | .0056666 .0182 0.31 0.756 -.0300049 .041338
ln_GDP_per | .0126002 .0118619 1.06 0.288 -.0106486 .035849
ln_emp_dens | .0042639 .0081688 0.52 0.602 -.0117467 .0202744
ln_FixedDI | -.0024622 .0076625 -0.32 0.748 -.0174803 .012556
ln_indu_str | .0062549 .0110288 0.57 0.571 -.0153612 .0278709
ln_tea_per | -.0316706 .0186299 -1.70 0.089 -.0681845 .0048434
ln_doc_per | -.0119927 .0096295 -1.25 0.213 -.0308662 .0068807
ln_gre_per | -.0134126 .0077927 -1.72 0.085 -.028686 .0018609
-------------+----------------------------------------------------------------
Wx |
ins_10k | -.1101112 .0235168 -4.68 0.000 -.1562033 -.0640191
ln_pop | -2.731433 .3513321 -7.77 0.000 -3.420031 -2.042834
ln_GDP_per | -1.191889 .0978764 -12.18 0.000 -1.383723 -1.000055
ln_emp_dens | 1.274961 .1232264 10.35 0.000 1.033441 1.51648
ln_FixedDI | .6932139 .0799528 8.67 0.000 .5365093 .8499184
ln_indu_str | 1.072119 .1197738 8.95 0.000 .8373665 1.306871
ln_tea_per | -.8538837 .3519525 -2.43 0.015 -1.543698 -.1640695
ln_doc_per | -.8204004 .218481 -3.76 0.000 -1.248615 -.3921856
ln_gre_per | -3.120411 .1957448 -15.94 0.000 -3.504063 -2.736758
-------------+----------------------------------------------------------------
Spatial |
rho | 4.617509 .1466779 31.48 0.000 4.330025 4.904992
-------------+----------------------------------------------------------------
Variance |
sigma2_e | .0074831 .0002096 35.70 0.000 .0070722 .0078939
-------------+----------------------------------------------------------------
SR_Direct |
ins_10k | .0239611 .0026483 9.05 0.000 .0187705 .0291517
ln_pop | -.0069616 .0179036 -0.39 0.697 -.042052 .0281289
ln_GDP_per | .0061032 .0118789 0.51 0.607 -.017179 .0293854
ln_emp_dens | .0118531 .0083588 1.42 0.156 -.00453 .0282361
ln_FixedDI | .0015804 .0075689 0.21 0.835 -.0132544 .0164153
ln_indu_str | .0122578 .0111952 1.09 0.274 -.0096844 .0342
ln_tea_per | -.0372955 .0186412 -2.00 0.045 -.0738315 -.0007595
ln_doc_per | -.0156879 .0097806 -1.60 0.109 -.0348575 .0034818
ln_gre_per | -.030305 .008717 -3.48 0.001 -.0473899 -.01322
-------------+----------------------------------------------------------------
SR_Indirect |
ins_10k | -.0000466 .0135823 -0.00 0.997 -.0266674 .0265742
ln_pop | 1.626941 .2388397 6.81 0.000 1.158824 2.095058
ln_GDP_per | .6929296 .0600374 11.54 0.000 .5752585 .8106007
ln_emp_dens | -.7904835 .0627765 -12.59 0.000 -.9135231 -.6674438
ln_FixedDI | -.4114165 .0545876 -7.54 0.000 -.5184062 -.3044268
ln_indu_str | -.6695471 .0670897 -9.98 0.000 -.8010404 -.5380538
ln_tea_per | .589073 .2206522 2.67 0.008 .1566027 1.021543
ln_doc_per | .5353626 .142433 3.76 0.000 .256199 .8145261
ln_gre_per | 1.927789 .1389189 13.88 0.000 1.655513 2.200065
-------------+----------------------------------------------------------------
SR_Total |
ins_10k | .0239145 .0129956 1.84 0.066 -.0015565 .0493855
ln_pop | 1.619979 .2373246 6.83 0.000 1.154831 2.085127
ln_GDP_per | .6990328 .0550801 12.69 0.000 .5910777 .8069878
ln_emp_dens | -.7786304 .0614533 -12.67 0.000 -.8990767 -.6581841
ln_FixedDI | -.4098361 .0535431 -7.65 0.000 -.5147787 -.3048934
ln_indu_str | -.6572893 .065499 -10.04 0.000 -.785665 -.5289136
ln_tea_per | .5517775 .215486 2.56 0.010 .1294326 .9741223
ln_doc_per | .5196747 .139795 3.72 0.000 .2456815 .7936679
ln_gre_per | 1.897484 .1395391 13.60 0.000 1.623993 2.170976
-------------+----------------------------------------------------------------
LR_Direct |
ins_10k | .0435471 .481839 0.09 0.928 -.90084 .9879341
ln_pop | .269298 5.940164 0.05 0.964 -11.37321 11.91181
ln_GDP_per | .174107 3.302126 0.05 0.958 -6.297942 6.646156
ln_emp_dens | -.1332378 3.459629 -0.04 0.969 -6.913986 6.64751
ln_FixedDI | -.086341 1.831853 -0.05 0.962 -3.676707 3.504025
ln_indu_str | -.0938687 2.608451 -0.04 0.971 -5.206339 5.018601
ln_tea_per | -.0288706 1.483939 -0.02 0.984 -2.937338 2.879596
ln_doc_per | .0890899 2.749492 0.03 0.974 -5.299816 5.477996
ln_gre_per | .2920785 7.724185 0.04 0.970 -14.84705 15.4312
-------------+----------------------------------------------------------------
LR_Indirect |
ins_10k | -.0597049 .3150391 -0.19 0.850 -.6771701 .5577604
ln_pop | .8777585 3.893841 0.23 0.822 -6.75403 8.509547
ln_GDP_per | .3244342 2.155578 0.15 0.880 -3.900421 4.54929
ln_emp_dens | -.4378991 2.262136 -0.19 0.847 -4.871604 3.995806
ln_FixedDI | -.2083978 1.195538 -0.17 0.862 -2.551609 2.134814
ln_indu_str | -.3859811 1.710477 -0.23 0.821 -3.738453 2.966491
ln_tea_per | .4466252 .9946368 0.45 0.653 -1.502827 2.396077
ln_doc_per | .3052174 1.795451 0.17 0.865 -3.213801 3.824236
ln_gre_per | 1.087565 5.056423 0.22 0.830 -8.822842 10.99797
-------------+----------------------------------------------------------------
LR_Total |
ins_10k | -.0161578 .1683809 -0.10 0.924 -.3461782 .3138626
ln_pop | 1.147057 2.074716 0.55 0.580 -2.919312 5.213425
ln_GDP_per | .4985412 1.155168 0.43 0.666 -1.765547 2.762629
ln_emp_dens | -.5711369 1.208754 -0.47 0.637 -2.940251 1.797978
ln_FixedDI | -.2947387 .6405193 -0.46 0.645 -1.550134 .9606561
ln_indu_str | -.4798497 .9091683 -0.53 0.598 -2.261787 1.302087
ln_tea_per | .4177546 .5380461 0.78 0.437 -.6367963 1.472306
ln_doc_per | .3943073 .9652189 0.41 0.683 -1.497487 2.286102
ln_gre_per | 1.379644 2.694185 0.51 0.609 -3.900863 6.66015
------------------------------------------------------------------------------
