You can find the spreadsheets for this video and some additional materials here: drive.google.com/drive/folders/1sP40IW0p0w5IETCgo464uhDFfdyR6rh7 Please consider supporting NEDL on Patreon: www.patreon.com/NEDLeducation
Hi sir,Great video.I have a quick question, suppose we use a series to test for stationarity using DF and ADF. We conclude that there is stationarity in the series.This means that variance of the series is constant right. And we use the same series to detect heterosedasticity in the series and we find that disturbances has no constant variance. How do we interpret this. The question open for everyone as I need to find the answer quickly. The question keeps me up at night😅
Thanks for the video. I need to do a white test for my university project. I have 15 stocks with their excess returns and I did regression for each one in excel. So, now I have their residuals and predicted y. I am not sure how can I do the white test. Is it correct to calculate the (predicted y)^2 and (residuals)^2. Then, run the regression for predicted y and (predicted y)^2 as independent variables and residuals square (residuals)^2 as the dependant variable. I would really appreciate it if you could help me :)
Hi Kian, and glad you liked the video! As for your question, the White test requires you to regress the squared residuals onto all independent variables, their squares, and cross-products. If you have just one independent variable (as it seems in your case), the White test would include two explanatory variables for the auxiliary regression: X and X^2. If you have got two explanatory variables, the auxiliary regression has five: X1, X2, X1^2, X2^2, and X1*X2; for three, it is nine: X1, X2, X3, X1^2, X2^2, X3^2, X1*X2, X1*X3, X2*X3, etc. You do not need to include the predicted Y in any way into the auxiliary regression. Hope it helps!
@@NEDLeducation i am testing right now , thanks! another question. For the Chi-sq stat is used in excel: CHISQ.DIST.RT in python i have used the stats.chi2.pdf as results i have half of the value that was calculated in excel, in this cas why should i multiply it by 2?
You can find the spreadsheets for this video and some additional materials here: drive.google.com/drive/folders/1sP40IW0p0w5IETCgo464uhDFfdyR6rh7
Please consider supporting NEDL on Patreon: www.patreon.com/NEDLeducation
Using the White test, we detect heteroskedasticity. How do we correct it?
Hi sir,Great video.I have a quick question, suppose we use a series to test for stationarity using DF and ADF. We conclude that there is stationarity in the series.This means that variance of the series is constant right. And we use the same series to detect heterosedasticity in the series and we find that disturbances has no constant variance. How do we interpret this. The question open for everyone as I need to find the answer quickly. The question keeps me up at night😅
Thanks for the video. I need to do a white test for my university project. I have 15 stocks with their excess returns and I did regression for each one in excel. So, now I have their residuals and predicted y. I am not sure how can I do the white test. Is it correct to calculate the (predicted y)^2 and (residuals)^2. Then, run the regression for predicted y and (predicted y)^2 as independent variables and residuals square (residuals)^2 as the dependant variable. I would really appreciate it if you could help me :)
Hi Kian, and glad you liked the video! As for your question, the White test requires you to regress the squared residuals onto all independent variables, their squares, and cross-products. If you have just one independent variable (as it seems in your case), the White test would include two explanatory variables for the auxiliary regression: X and X^2. If you have got two explanatory variables, the auxiliary regression has five: X1, X2, X1^2, X2^2, and X1*X2; for three, it is nine: X1, X2, X3, X1^2, X2^2, X3^2, X1*X2, X1*X3, X2*X3, etc. You do not need to include the predicted Y in any way into the auxiliary regression. Hope it helps!
@@NEDLeducation thank you for the reply
Hi once more xD
does: scipy.stats.f.sf() does the same as F.DIST.RT( in excel?
regards
Hi Antonio, yes, indeed, you can use this function to run the F-test in Python.
@@NEDLeducation i am testing right now , thanks!
another question.
For the Chi-sq stat
is used in excel: CHISQ.DIST.RT
in python i have used the stats.chi2.pdf as results i have half of the value that was calculated in excel, in this cas why should i multiply it by 2?
@@antoniofernandes5751 For the chi-squared right-tailed test, you can use scipy.stats.chi2.sf in a similar manner.
@@NEDLeducation perfetc, thanks once more!