Probit model, outcome does not vary

[Aaron Nagel]

Hello everybody,

I used predicted variables from PCA for an EFA and want to implement my findings (11 factors) in a probit model.

What I did so far and what is planned:

1. I cut my dataset into 4 sets
2. I used PCA to trim down from my ~180 variables (I now have 48 items and 9 components describing those)
3. I used these 9 components and implemented them into my second set for an EFA
4. After finding the underlying structure of 11 factors, I want to implement these findings on a third set to regress a probit model
5. The last set of the 4 is for running the probit model.

Now:
I am somewhat stuck on how to implement the probit model.
I did use "predict fa1 fa2 ... fa11" to get the new variables from my factor I found via the FA. Through "mkmat...., mat(probitraw) obs nchar(1)", "mat probitfa = probitraw*fa" and "svmat probitfa, names( col )" I was able to implement the structure / factors onto my new set 3.

Now running probit on the list of variables found via this, plus some extra dummy variables, seems to have a problem.

It shows:

outcome does not vary; remember:
0 = negative outcome,
all other nonmissing values = positive outcome

From my research it apparently comes from an incorrect implementation of binary variables, but they seem to be fine.

Here my results from my "summarize $probitlist"

Variable	Obs	Mean	Std. Dev.	Min	Max
size_fin_fa	456	2.39e+07	9.67e+07	64475.97	1.10e+09
impactpwor~a	456	-304119.8	7107358	-1.10e+08	5.35e+07
opexratio_fa	456	-13441.49	528577.8	-4858843	7785143
nparatio_fa	456	-4174440	1.80e+07	-1.90e+08	-12590.47
size_emp_fa	456	2403224	1.40e+07	-3218961	2.23e+08
risk_fa	456	-3355884	1.65e+07	-2.56e+08	408332
opincmargi~a	456	-2767453	1.30e+07	-1.80e+08	2.62e+07
nonintincm~a	456	-435768.6	5186972	-8.96e+07	2.36e+07
netloanrat~a	456	4631591	1.80e+07	-1125717	1.72e+08
netincrati~a	456	6312471	2.77e+07	-1.11e+08	3.19e+08
roe_fa	456	-538707.7	1.14e+07	-1.03e+08	1.37e+08
usa	456	.9166667	.276689	0	1
bank_based	456	.0526316	.2235421	0	1
market_based	456	.004386	.0661538	0	1
eastern_eu~e	456	.0153509	.123079	0	1
outliers	456	.002193	.0468293	0	1
interconti~l	456	.0065789	.0809322	0	1
bank_type	456	.0504386	.2190886	0	1

Can someone help?

Thank you in advance!


Aaron

[Clyde Schechter]

I don't think anybody can help you with this unless you post back with:

a) an excerpt from the data set you are trying to do the probit regression on (use -dataex-) If you are running version 16 or a fully updated version 15.1 or 14.2, -dataex- is already part of your official Stata installation. If not, run -ssc install dataex- to get it. Either way, run -help dataex- to read the simple instructions for using it. -dataex- will save you time; it is easier and quicker than typing out tables. It includes complete information about aspects of the data that are often critical to answering your question but cannot be seen from tabular displays or screenshots. It also makes it possible for those who want to help you to create a faithful representation of your example to try out their code, which in turn makes it more likely that their answer will actually work in your data.

b) the actual -probit- command you tried to run.

[Richard Williams]

To Clyde's excellent advice, I will add that, after reading your message 3 or 4 times, I still could not figure out what the DV is. It should be coded 0/1. If it is instead coded 1/2, you will get error messages like you got.

If your DV is indeed coded 0/1, it could be that missing data on the independent variables is wiping out either all the 0s or all the 1s on your dv.

So, the most basic thing to do is run a frequency on your DV. It may be that all you need to do is recode it. But if it is more complicated than that we will need more information.

[Aaron Nagel]

Thank you a lot, sadly I had to go back to my data to fix collinearity issues. I will make a separate entry, since it does not fit to this question here.