Heteroskedasticity
6 important questions on Heteroskedasticity
What are the consequences of heteroskedasticity?
It does cause bias in variance (MLR5), so t statistic, F statistic, standard errors and confidence intervals are invalid.
Are OLS estimators efficient under heteroskedasticity?
What is the difference between heteroskedasticity-robust inference and t statistic / F statistic of chapter 4?
- Higher grades + faster learning
- Never study anything twice
- 100% sure, 100% understanding
How does the Breusch-Pagan test work?
Alternative hypothesis (violation of MLR5): var(u|x1,xk)=h(x1,..,xk)o2 with h(x1,..,xk)=exp(...)
Because of MLR4, var (u|x1,..,xk)=E(u2|x1,..,xk) gives H0: E[u2|x1,..,xk]=o2
Assume a linear relation u2=delta0+delta1x1+... With E[v|x1,..,xk]=0
Then we can use F statistic to test H0:delta1=deltak=0
What is the motivation behind weighted least squares (WLS)?
How do we derive WLS from OLS?
The question on the page originate from the summary of the following study material:
- A unique study and practice tool
- Never study anything twice again
- Get the grades you hope for
- 100% sure, 100% understanding
![Trustpilot Logo](/images/design_2020/lp/clp-trustpilot-en.png)