Linear Regression - Robust standard error
Posted: Thu Apr 01, 2021 8:57 pm
Dear Jamovi team,
Thank you for making jamovi available. I teach statistics to economics, business and international relations students and we've been using for (way too many) years SPSS at the undergraduate level.
I recently heard about jamovi and I am very interested in replacing SPSS by jamovi in my classes. However, in regression analysis, we use the robust standard error of the coefficients (in fact, the White-Huber estimator for standard error) a lot. This is a very importance feature for economics, business and international relations students as the robust standard error corrects for misspecifications and deals with heteroskedasticity.
While this is available in SPSS (in GLM), I couldn't find it in jamovi. In jamovi, we have the possibility of testing for homoskedasticity (with the moretests module), but it would be relevant to apply the robust estimator in case the test points out for the existence of heteroskedasticity.
Could you please help me and tell me how to find such a feature? In the event jamovi does not include that, could you please tell me if you plan to include it in next versions (because I am preparing the fall semester and would like to use jamovi)?
Thank you
Thank you for making jamovi available. I teach statistics to economics, business and international relations students and we've been using for (way too many) years SPSS at the undergraduate level.
I recently heard about jamovi and I am very interested in replacing SPSS by jamovi in my classes. However, in regression analysis, we use the robust standard error of the coefficients (in fact, the White-Huber estimator for standard error) a lot. This is a very importance feature for economics, business and international relations students as the robust standard error corrects for misspecifications and deals with heteroskedasticity.
While this is available in SPSS (in GLM), I couldn't find it in jamovi. In jamovi, we have the possibility of testing for homoskedasticity (with the moretests module), but it would be relevant to apply the robust estimator in case the test points out for the existence of heteroskedasticity.
Could you please help me and tell me how to find such a feature? In the event jamovi does not include that, could you please tell me if you plan to include it in next versions (because I am preparing the fall semester and would like to use jamovi)?
Thank you