Based on the below result, I have the following questions
Code:
reghdfe lever_w /// > i.ibc##i.treat2 /// > size_w nfa_ta_w cash_ta_w trade_credit_w sales_grow_w roa_w pb_w cfo_ta_w rddcc_dum div_dum nw_ta_w age , absorb > (ff48##i.year code) cluster (code) (dropped 43 singleton observations) note: 1bn.ibc is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = 1.0 > e-09) note: 1bn.treat2 is probably collinear with the fixed effects (all partialled-out values are close to zero; tol = > 1.0e-09) (MWFE estimator converged in 10 iterations) note: 1.ibc omitted because of collinearity note: 1.treat2 omitted because of collinearity HDFE Linear regression Number of obs = 4,559 Absorbing 2 HDFE groups F( 13, 744) = 32.69 Statistics robust to heteroskedasticity Prob > F = 0.0000 R-squared = 0.9216 Adj R-squared = 0.8997 Within R-sq. = 0.5170 Number of clusters (code) = 745Root MSE = 0.0650 (Std. err. adjusted for 745 clusters in code) -------------------------------------------------------------------------------- | Robust lever_w | Coefficient std. err. t P>|t| [95% conf. interval] ---------------+---------------------------------------------------------------- 1.ibc | 0 (omitted) 1.treat2 | 0 (omitted) | ibc#treat2 | 1 1 | .0109666 .0077777 1.41 0.159 -.0043022 .0262354 | size_w | -.004929 .0099703 -0.49 0.621 -.0245022 .0146442 nfa_ta_w | .1371436 .0258866 5.30 0.000 .086324 .1879631 cash_ta_w | .0873572 .0527998 1.65 0.098 -.0162971 .1910114 trade_credit_w | .204013 .0421171 4.84 0.000 .1213304 .2866955 sales_grow_w | -.0072218 .003361 -2.15 0.032 -.01382 -.0006235 roa_w | -.0428388 .0298777 -1.43 0.152 -.1014933 .0158158 pb_w | -5.14e-06 .0011595 -0.00 0.996 -.0022813 .0022711 cfo_ta_w | -.1118 .0203783 -5.49 0.000 -.1518058 -.0717941 rddcc_dum | -.0096273 .0058958 -1.63 0.103 -.0212018 .0019471 div_dum | -.0049204 .0053542 -0.92 0.358 -.0154317 .0055908 nw_ta_w | -.5787685 .0397092 -14.58 0.000 -.656724 -.500813 age | -.0145416 .0329906 -0.44 0.660 -.0793074 .0502242 _cons | .5850335 .1386969 4.22 0.000 .3127495 .8573174 -------------------------------------------------------------------------------- Absorbed degrees of freedom: ------------------------------------------------------------+ Absorbed FE | Categories - Redundant = Num. Coefs | --------------------+---------------------------------------| ff48#year | 239 0 239 | code | 745 745 0 *| ------------------------------------------------------------+ * = FE nested within cluster; treated as redundant for DoF computation
I have seen the use of -margins-command and I used it here without much knowledge and here are my outputs
Code:
margins treat2#ibc, noestimcheck
Predictive margins Number of obs = 4,559
Model VCE: Robust
Expression: Linear prediction, predict()
------------------------------------------------------------------------------
| Delta-method
| Margin std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
treat2#ibc |
0 0 | .300477 .0015934 188.58 0.000 .297354 .3036
0 1 | .300477 .0015934 188.58 0.000 .297354 .3036
1 0 | .300477 .0015934 188.58 0.000 .297354 .3036
1 1 | .3114435 .0061843 50.36 0.000 .2993226 .3235645
------------------------------------------------------------------------------
. margins treat2, dydx(ibc) noestimcheck
Average marginal effects Number of obs = 4,559
Model VCE: Robust
Expression: Linear prediction, predict()
dy/dx wrt: 1.ibc
------------------------------------------------------------------------------
| Delta-method
| dy/dx std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
0.ibc | (base outcome)
-------------+----------------------------------------------------------------
1.ibc |
treat2 |
0 | 0 (omitted)
1 | .0109666 .0077777 1.41 0.159 -.0042774 .0262105
------------------------------------------------------------------------------
Note: dy/dx for factor levels is the discrete change from the base level.
. margins ibc, dydx(treat2), noestimcheck
invalid 'noestimcheck'
r(198);
Also why margins ibc, dydx(treat2), noestimcheck didnt work
0 Response to Interpreting an Statistically Insignifcant interaction of a classical DID analysis
Post a Comment