I tried to get the estimated coefficients from the relogit model of the Zelig package, but R returns NULL in the console when I run the script. Let's see the following example taken from http://ftp.uni-bayreuth.de/math/statlib/R/CRAN/doc/vignettes/Zelig/relogit.pdf :



z.out1 <- zelig(conflict ~ major + contig + power + maxdem + mindem + years, 
            data = mid, model = "relogit", tau = 1042/303772) 


I got the following output:

    Call:  relogit(formula = form, bias.correct = bias.correct, case.control = case.control, 
    tau = tau, data = data)

             Estimate     Std. Error   z value      Pr(>|z|)   
(Intercept)   -7.508e+00    1.793e-01   -4.188e+01   1.535e-304
major          2.432e+00    1.572e-01    1.547e+01    4.556e-52
contig         4.108e+00    1.573e-01    2.612e+01   3.852e-136
power          1.054e+00    2.168e-01    4.861e+00    1.228e-06
maxdem         4.804e-02    1.004e-02    4.783e+00    1.805e-06
mindem        -6.413e-02    1.277e-02   -5.020e+00    5.444e-07
years         -6.293e-02    5.692e-03   -1.106e+01    6.687e-28

Degrees of Freedom: 3125 Total (i.e. Null);  3119 Residual
Null Deviance:      3979 
Residual Deviance: 1869     AIC: 1883

Prior correction performed with tau = 0.003430204 
Rare events bias correction performed
> z.out1$coefficients

Then I tried to run the following script from the same model, following http://docs.zeligproject.org/en/latest/zelig-relogit.html :


but I got the same NULL answer from the console. How I can get the coefficient values vector?


It should be z.out1$getcoef(), getcoef() is a function call. It will return a list, the coefficient vector is thus z.out1$getcoef()[[1]].

The normal way, i.e. coef(summary(MODEL)) as we will do for glm etc, doesn't work for Zelig.


Not the answer you're looking for? Browse other questions tagged or ask your own question.