Stata Tutorial: F-statistic for joint significance testing

[@justsomeone9394]

I was looking for such an explanation all over youtube. As said by others, clear and concise. Thanks

[@TheonlyIsaac07]

Great stuff Mike, I like how you recapped the maths before showing the STATA commands. Thanks.

[@paraskapoor7791]

This was really good, the concept is completely clear for me

[@mikejonaseconometrics1886]

Thank you - glad it was helpful!

[@yuyaoyao.]

Super helpful for understanding how to test joint hypothesis! Thx a lot!

[@mikejonaseconometrics1886]

Glad it was helpful!

[@javaid607]

You legend. God bless you. Love all your stata videos.

[@numangun7172]

This was clear and concise. Thank you for the video!

[@nicholash6359]

Amazing and easy to understand, thank you! Just what I was looking for

[@mikejonaseconometrics1886]

You're very welcome!

[@aminegomri1403]

I would like to thank you for your precious and cristal clear work! Amazing help

[@weijieyang8383]

Very helpful and clear interpretation!

[@richardsalinas1986]

thx! amazing! grettings from Perú

[@felixparis2409]

thanks for saving with my coursework

[@mikejonaseconometrics1886]

You’re welcome - glad it helped!

[@yuceler77]

Another great video. Thanks.

[@frickinfrick8488]

Love you man, very helpful

[@clarketoby2]

Great video, big help

[@sbhusan212]

Please help with the Wald Test in Stata for structural break

[@ebadhashemi5970]

very clear explanation!

[@villageandanimals6986]

Thanks bro! keep it up

[@maximekerstholt6107]

Hi! Thanks for this video. I really want to know how I can do this in SPSS after I made up a GLMM. I want to know the joint significance of some variables, but don't know how to do it in SPSS. Can you help me with that?

[@sadafsarrafi3289]

Thanks. It was so useful

[@miguelgiraldo1938]

Excellent explanation! what it’s the meaning of q next to the F ?

[@loanie666]

Hi, is it possible to test the joint significance of the intercept = 3 and beta1 = -2 (for example) ? How do you test an intercept with the "test" command ?

[@mikejonaseconometrics1886]

You can combine two test commands with the 'accumulate' option: "test _cons=3" "test beta1=-2, accumulate" The resulting F-test will include both restrictions. Interesting question!

[@loanie666]

@@mikejonaseconometrics1886 Thank you very much!

[@tanyadyne]

If I fail to reject the null hypothesis (which in my case, is linearity), does that mean I can conclude that my restricted model is more valid/better than the unrestricted model?

[@RavindraMulye]

Sir Is there a way to resolve stata r(950) error...

[@guyfawkes7949]

5.12 > 2.30. Where do we get the 2.30 value from?

[@danwood5027]

Nice and clear video! I am wondering though, I have it where the p-value is 0.4322, meaning we fail to reject the null hypothesis. However when I do test variable1 variable2 I get the exact same results. Does this mean the variables are jointly statistically significant or not? How does the p-value come into it?

[@mikejonaseconometrics1886]

Dan, The p-value will give you the level of significance at which you can just exactly reject the null. So regardless of the test, a p-value greater than 0.10 indicates that the calculated test statistic does not surpass the 10% critical value.

[@tallyskalynkafeldens1753]

So great!!!!

[@nizarneffati9675]

Thanks a lot. Just a little question : If we want to test (beta1+beta2)=0, how to do that in stata ?

[@mikejonaseconometrics1886]

You can use the F-test here as well. If your regression is 'regress y x1 x2', the test command will simply be 'test x1+x2=0'

[@annaleeburks1230]

Thank you so much!

[@nortongartino4602]

Thank you so much

[@EmperorDraconianIV]

how do you do this for mlogit

[@BLINKBOXHD]

if you get a p value of 0, does that mean there is an error or that those variables have high joint significance?

[@mikejonaseconometrics1886]

It is not an error. The p-value may appear to be zero, but if you go out enough decimal points you will always see a positive value. It means you can reject the null at well beyond 1% significance.

[@yiningzhang5891]

It's so helpful!!!!!!!!! thxxxxx

[@juliomillan9263]

thanks a lot!

[@mikejonaseconometrics1886]

You’re welcome - hope it was helpful!

[@arianajimena0803]

Lo quiero mucho sr que explica stata

[@omar4901]

Thank you!!!

[@daniele2761]

are you assuming homoskedasticity? Wouldn't it be more correct to write the regression model with " , r " in the end before doing the F test?

[@mikejonaseconometrics1886]

Yes, the implicit assumption is a homoskedastic error, but I should have mentioned it, - thank you! I try to focus on one task at a time for the purposes of the tutorial, but in reality, there are many issues that have to be dealt with before we can 'believe' the results.

[@zoozolplexOne]

Cool !!

[@jeremy8509]

thanks a lot !!!