I am trying to graph my results using the margins command, however, it returns the "not estimable" error. I have tried the advice here (https://www.statalist.org/forums/for...-not-estimable) but it does not work for me.
Code:
areg C1 i.sonbirth##i.year [pweight= panel_wt_06_12_18] , absorb(Findid) (sum of wgt is 7,684.09828691185) note: 1.sonbirth omitted because of collinearity Linear regression, absorbing indicators Number of obs = 8,535 Absorbed variable: Findid No. of categories = 3,540 F( 4, 4991) = 52.86 Prob > F = 0.0000 R-squared = 0.5036 Adj R-squared = 0.1513 Root MSE = 1.3278 ------------------------------------------------------------------------------- | Robust C1 | Coef. Std. Err. t P>|t| [95% Conf. Interval] --------------+---------------------------------------------------------------- 1.sonbirth | 0 (omitted) | year | 1 | -.2650297 .1069619 -2.48 0.013 -.4747219 -.0553374 2 | .4925483 .1156093 4.26 0.000 .2659033 .7191934 | sonbirth#year | 1 1 | .2409295 .1152284 2.09 0.037 .0150313 .4668278 1 2 | .0648046 .1315297 0.49 0.622 -.1930513 .3226606 | _cons | -.0330636 .0355901 -0.93 0.353 -.1028359 .0367087 ------------------------------------------------------------------------------- . margins sonbirth##year Predictive margins Number of obs = 8,535 Model VCE : Robust Expression : Linear prediction, predict() ------------------------------------------------------------------------------- | Delta-method | Margin Std. Err. t P>|t| [95% Conf. Interval] --------------+---------------------------------------------------------------- sonbirth | 0 | . (not estimable) 1 | . (not estimable) | year | 0 | -.0330636 .0355901 -0.93 0.353 -.1028359 .0367087 1 | -.12697 .0259481 -4.89 0.000 -.1778397 -.0761003 2 | .505513 .0400142 12.63 0.000 .4270676 .5839585 | sonbirth#year | 0 0 | . (not estimable) 0 1 | . (not estimable) 0 2 | . (not estimable) 1 0 | . (not estimable) 1 1 | . (not estimable) 1 2 | . (not estimable) -------------------------------------------------------------------------------
0 Response to margins "not estimable"
Post a Comment