Hello all,
I have calculated -xtlogit,re- for my unbalanced panel (in Stata 14.2) because my dependent variable has only the expressions 1 and 0. Then, I summarized the results of the regressions in a table using the user-written command -estout-. You can see this table at the end of this post.
My question is about the interaction results in this table. I don't understand why for the interactions between vGP and GP as well as between vRC and CC a separate value is output for the case when vGP or vRC equals 1 and once for the case when vGP or vRC equals 0.
vGP and vRC are both binary variables. For vGP, the 1 represents "National" and the 0 represents "International". For vRC, the 1 stands for "Existing previous experience" and the 0 for "Non-existing previous experience". So far I can understand everything, but I have never seen such a presentation of the regression results. I just know it in such a way that in papers exclusively the values are indicated, if vGP or vRC are equal to 1. The values for the case when vGP or vRC are equal to 0 are never output. Can someone kindly help me with this and explain to me why these values are shown and what I can do so that they do not appear in the regression output?
Many, many thanks for the help and support.
I have calculated -xtlogit,re- for my unbalanced panel (in Stata 14.2) because my dependent variable has only the expressions 1 and 0. Then, I summarized the results of the regressions in a table using the user-written command -estout-. You can see this table at the end of this post.
My question is about the interaction results in this table. I don't understand why for the interactions between vGP and GP as well as between vRC and CC a separate value is output for the case when vGP or vRC equals 1 and once for the case when vGP or vRC equals 0.
vGP and vRC are both binary variables. For vGP, the 1 represents "National" and the 0 represents "International". For vRC, the 1 stands for "Existing previous experience" and the 0 for "Non-existing previous experience". So far I can understand everything, but I have never seen such a presentation of the regression results. I just know it in such a way that in papers exclusively the values are indicated, if vGP or vRC are equal to 1. The values for the case when vGP or vRC are equal to 0 are never output. Can someone kindly help me with this and explain to me why these values are shown and what I can do so that they do not appear in the regression output?
Many, many thanks for the help and support.
Code:
quietly xtlogit SE c.log_Age c.SC c.CS1 SL, re vce(cluster pid) or nolog
estimates store xtlogit_re_c_1
quietly xtlogit SE vGP c.vCS vRC c.log_Age c.SC c.CS1 SL, re vce(cluster pid) or nolog
estimates store xtlogit_re_c_2
quietly xtlogit SE vGP c.vCS vRC c.GP c.CS c.CC c.log_Age c.SC c.CS1 SL, re vce(cluster pid) or nolog
estimates store xtlogit_re_c_3
quietly xtlogit SE vGP c.vCS c.GP vRC c.CS c.CC vGP#c.GP c.log_Age c.SC c.CS1 SL, re vce(cluster pid) or nolog
estimates store xtlogit_re_c_4
quietly xtlogit SE vGP c.vCS vRC c.GP c.CS c.CC c.vCS#c.CS c.log_Age c.SC c.CS1 SL, re vce(cluster pid) or nolog
estimates store xtlogit_re_c_5
quietly xtlogit SE vGP c.vCS vRC c.GP c.CS c.CC vRC#c.CC c.log_Age c.SC c.CS1 SL, re vce(cluster pid) or nolog
estimates store xtlogit_re_c_6
quietly xtlogit SE vGP c.vCS vRC c.GP c.CS c.CC vGP#c.GP c.vCS#c.CS vRC#c.CC c.log_Age c.SC c.CS1 SL, re vce(cluster pid) or nolog
estimates store xtlogit_re_c_7
estout xtlogit_re_c_1 xtlogit_re_c_2 xtlogit_re_c_3 xtlogit_re_c_4 xtlogit_re_c_5 xtlogit_re_c_6 xtlogit_re_c_7 using Main-Regression_1.csv, cells(b(star fmt(3) label(Odds ratio)) se(par fmt(2) label(std.errors))) varwidth(23) starl(+ 0.1 * 0.05 ** 0.01 *** 0.001) eform stats(N df_m ll chi2, labels ("No. of obs." "Degrees of freedom" "Log likelihood" "Wald chi2") fmt(3 0 1)) label legend varlabels(_cons Constant)
Comment