Hello, I am trying to estimate the number of correct predictions in a linear model using robust and a linear model with wls. My instructor showed doing this regress diab owgt obese exer cig alc inc coll marr male age, robust
predict probl, xb
generate cdiab=(probl >=0.5 & diab==1)
generate cnodiab=(probl <0.5 & diab==0)
generate correct =(cdiab==1 | cnodiab==1)
sum diab cdiab cnodiab correct
but is this specific for either model? And it only prints out this:
Variable | Obs Mean Std. Dev. Min Max
-------------+---------------------------------------------------------
diab | 5,051 .1282914 .3344471 0 1
cdiab | 5,051 0 0 0 0
cnodiab | 5,051 .8717086 .3344471 0 1
correct | 5,051 .8717086 .3344471 0 1
It does not include a table that shows the predictions as the instructor said it would. Can anyone tell me if the information I need is hidden in here or if I did something wrong?
Thank you
Related Posts with percent of correct predictions
Color of Confidence IntervalsDear Stata users, Is it possible to have the same color for 95 CI in all groups? I am using the fol…
Use of pstolerance with teffects aipwHi everyone, I'd like to use the IPW estimator in my model, with treatment effects. The treatment i…
Panel data - linking individuals across yearsDear All, I have data from household panel survey in a long form (several rows for one individual f…
ivpoisson and gmmDear all, Is it feasible to adopt the GMM approach to solve endogeneity issues in gravity modelling…
Choosing between better estimates and the observation numberHello dear Stata Community, I have a general question about estimators and the number of observatio…
Subscribe to:
Post Comments (Atom)
0 Response to percent of correct predictions
Post a Comment