Compare: Logit regress with Categorical variables with Chi2 test

[ThongKe Club]

Dear all,
I have problem with Logit regress.

use https://thongke.club/data/logit.dta 

logit Y GioiTinh Tuoi HocVan KinhNghiem DienTich 

GioiTinh (0;1) is Sig (10%)

But when I test by Chi2



I have P-value = 0.911 ( No Sig).

Please help me, why ?

Thank you very much !

[Carlo Lazzaro]

ThongKe:
there's nothing surprising with your results, as the coefficients of -logit- are adjusted for the remaining predictors, whereas it does not happen with -tab-.

[ThongKe Club]

Hey Carlo,
How to use margins comand to explain for Categorical variables

I only know two command to show margins

margins, dydx(*) 

mfx 

But two command dont support for Categorical variables.

Thanks,

ThongKe:
there's nothing surprising with your results, as the coefficients of -logit- are adjusted for the remaining predictors, whereas it does not happen with -tab-.

[Carlo Lazzaro]

ThongKe:
if you take a look at -help margins- (that replaces -mfx-) and related entry in Stata .pdf manual, you can see how to deal with categorical variables.

[ThongKe Club]

Carlo:
i use this command: logit Y i.GioiTinh Tuoi HocVan KinhNghiem DienTich to do that.
And margins, dydx(*)

That's right or wrong Thanks,

[Carlo Lazzaro]

ThongKe:
try:

margins GioiTinh

That said, it's not up to interested listers to reply to yes/no questions, but to the original poster to find out a solution for her/his own problem following the advice provided.
In your case you could have easily discovered yourself that categorical variables do support -margins- just taking a look at Example 2, -margins- entry, Stata .pdf manual.

[ThongKe Club]

Thanks for your support, Carlo

ThongKe:
try:

margins GioiTinh

That said, it's not up to interested listers to reply to yes/no questions, but to the original poster to find out a solution for her/his own problem following the advice provided.
In your case you could have easily discovered yourself that categorical variables do support -margins- just taking a look at Example 2, -margins- entry, Stata .pdf manual.

[William Lisowski]

ThongKe Club

Along with the documentation for the margins command, you may find the explanations you need, and more, in the nice overview of margins prepared by Richard Williams, a frequent contributor here, at https://www3.nd.edu/~rwilliam/xsoc73994/Margins01.pdf with a more detailed paper in the Stata Journal at http://www.stata-journal.com/article...article=st0260. I'll also note that Margins01.pdf is followed by Margins02.pdf ... Margins05.pdf covering more specialized topics.

And all five of these PDFs, and plenty more of use to someone learning about the analysis of categorical data, are linked to from https://www3.nd.edu/~rwilliam/xsoc73994/index.html - the material mentioned above is found in the section headed Interpreting results: Adjusted Predictions and Marginal effects.

---

With that said, it takes little effort to see that your "name" is the URL of a commercial web site that (with proper diacritical marks) Google will translate as "statistical club" rather than a personal name, as members are requested to provide. Your avatar photo is of a popular internet personality who seems unlikely to be interested in statistical analysis at this stage of her academic career. Your profile includes the "field" that Google Translate tells me is in English "prestigious economic dissertation site" which may discourage members of Statalist from helping, as the result seems to abet academic fraud.

[Carlo Lazzaro]

Thanks William for pointing this out.
Very disappointing indeed: I should have been warned about something weird just from tones and manners embedded in the previous post, but I thought that they came from someone frustrated by repeated unsuccessful attempts.
For what it worths, I will not consider any further query from this anonymous poster.

[ThongKe Club]

Thank you for your sharing and support, William.

Should not believe in google search. I just a research and don't want to show my avatar. All of us here just for academic research. And I am interesting in this filed.
I am so disappointed to hear from William.

In my way, I always want to share my knowledge and my found with everyone in this forum and our web.

In my point, all of us learn from each other to improve our skill.

Today is a bad Friday!

Best regards,

ThongKe Club

Along with the documentation for the margins command, you may find the explanations you need, and more, in the nice overview of margins prepared by Richard Williams, a frequent contributor here, at https://www3.nd.edu/~rwilliam/xsoc73994/Margins01.pdf with a more detailed paper in the Stata Journal at http://www.stata-journal.com/article...article=st0260. I'll also note that Margins01.pdf is followed by Margins02.pdf ... Margins05.pdf covering more specialized topics.

And all five of these PDFs, and plenty more of use to someone learning about the analysis of categorical data, are linked to from https://www3.nd.edu/~rwilliam/xsoc73994/index.html - the material mentioned above is found in the section headed Interpreting results: Adjusted Predictions and Marginal effects.

---

With that said, it takes little effort to see that your "name" is the URL of a commercial web site that (with proper diacritical marks) Google will translate as "statistical club" rather than a personal name, as members are requested to provide. Your avatar photo is of a popular internet personality who seems unlikely to be interested in statistical analysis at this stage of her academic career. Your profile includes the "field" that Google Translate tells me is in English "prestigious economic dissertation site" which may discourage members of Statalist from helping, as the result seems to abet academic fraud.

[Phil Bromiley]

Let me try to explain without making you feel bad. We on this list serve use our real names as a matter of honesty. You're asking for our help so it behooves you to tell us who you are. We also generally follow the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. As is generally true in email as well, bold and large fonts are to be avoided - they are like shouting.

This list serve works on folks being willing to help others (who they do not know and probably will never meet) without any concrete reward. It is expected that those asking for help will follow the FAQ. That you try to hide your identity makes your posting suspicious - one of the rules here is that we don't do folks homework for them.