probit regression

[jan dhoore]

When conducting the probit regression, i get multiple omitted variables. How do I solve this? By regressing each omitted variable as a dependent variables or is there an other way to account for these omitted variables?

Also when I try to export this stata output into an excel file using / outreg2 variables, replace e(all) label excel alpha(0.001, 0.01, 0.05) symbol(***, **, *), i get an error saying "option e() not allowed". How can I solve this problem?

Thanks in advance!!

[William Lisowski]

Welcome to Statalist.

Without seeing your probit command and its output, it's difficult to be very helpful. And it's not clear what you mean by "solving" the problem.

You've apparently already seen the advice in the Stata FAQ at

https://www.stata.com/support/faqs/s...ted-variables/

based on your proposal to regress omitted variables as dependent variables. The result of doing this tells you what variables each omitted variable is collinear with, perhaps, but there's no way to get the omitted variable back into the model without excluding some other variable, so I don't know that this is a "solution" of the sort you are looking for.

In general, the most common reason for omitted variables is inclusion of a group of indicator variables ("dummy variables") that cover every category. For example, if everyone in your population has a gender of either female or male, and you have separate indicator variables for both female and male, then you have to exclude either the female or the male indicator variable, since together they add up to 1, which is collinear with the constant term. This is often called the "dummy variable trap" and is discussed in Wikipedia at the following location,

https://en.wikipedia.org/wiki/Dummy_...my_independent

Again, there's no "solution" to this.

For a more thorough discussion, should these pointers not prove helpful, you might take a few moments to review the Statalist FAQ linked to from the top of the page, as well as from the Advice on Posting link on the page you used to create your post. Note especially sections 9-12 on how to best pose your question. It's particularly helpful to copy commands and output from your Stata Results window and paste them into your Statalist post using code delimiters [CODE] and [/CODE], as described in section 12 of the FAQ.

The more you help others understand your problem, the more likely others are to be able to help you solve your problem.

[jan dhoore]

Okay, thanks for the information I'm new on this platform so i dont reqlly know the usual interaction. However my data is present below, maybe not in the way it's best readable, but i think it is still cleqr to read. As you can see, I have three omitted variables (all of them are dummies). All of these variables make up for the completness of a white paper, So it is logical the ommited variables are highly correlated with the variable complete white paper. For every problem, there was a solution as well, so this correlation is 1 and. Furthermore, both are found back in the whole sample so it makes sense that they are omitted. However, financial details are not always found back but is still omitted in my output with makes it kind of strange for me. So if I understand you well I have to delete all omitted variables except for one, or am I wrong?

----------------------+----------------------------------------------------------------
tokensdistributed | -1.824454 .8578954 -2.13 0.033 -3.505898 -.1430096
hardcaplog | .1131385 .1253064 0.90 0.367 -.1324576 .3587345
Patentactivitydummy | .3764257 .3852203 0.98 0.328 -.3785922 1.131444
Completewhitepaper | -.5022828 .3432484 -1.46 0.143 -1.175037 .1704717
Problemdummy | 0 (omitted)
Solutiondummy | 0 (omitted)
Technicalaspectdummy | .5402596 .6296035 0.86 0.391 -.6937406 1.77426
Financialdetailsdummy | 0 (omitted)
GitHubdummy | .5708313 .414692 1.38 0.169 -.2419502 1.383613
numberoftotalcommits | .0000674 .0000442 1.52 0.128 -.0000193 .0001541
Diversityindex | 1.150011 1.88543 0.61 0.542 -2.545365 4.845386
utilitydummy | .6478353 .6807837 0.95 0.341 -.6864763 1.982147
Tokensupplylog | -.0149982 .0711659 -0.21 0.833 -.1544809 .1244844
durationindays | -.0018485 .0033664 -0.55 0.583 -.0084466 .0047496
Country_n | .0348053 .0202925 1.72 0.086 -.0049673 .0745779
WPpages | .0042897 .0129377 0.33 0.740 -.0210678 .0296471
_cons | -1.142926 2.110682 -0.54 0.588 -5.279786 2.993934]

Thanks in advance

[Phil Bromiley]

Jan,
You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output (fixed spacing fonts help), and sample data using dataex. Assume we are not from your area. We don't even know what the columns are in the results you posted or what estimator (I assume an estimator) produced the results.

As William pointed out, there is probably a problem that your dummies add to one. Try regressing one of the dummies (say ProblemDummy) on the other rhs variables to try to understand what is going on .