# What would be the Variance of Intercept in case of heteroskedasticity? [closed]

Question is regarding Regression Analysis. what is the variance of intercept in case of heteroskedasticity?

-
To answer this, you need a quantitative model for how the error variance behaves. Which one do you have in mind? – whuber Sep 4 '12 at 18:42
Linear Model y=a+b1x1+...bkxk+error – Bioinformatics Sep 4 '12 at 19:23
Right, but that model is not a model of heteroscedasticity. You have to specify how the variance of "error" depends on the other data. – whuber Sep 4 '12 at 19:29
variance = sqrt(predictors) – Bioinformatics Sep 5 '12 at 5:51
This makes no sense with multiple predictors (as indicated in your first comment): which square root should one use? – whuber Sep 5 '12 at 12:39

## closed as not a real question by Macro, gung, Andy W, Peter Flom, Rob HyndmanSep 30 '12 at 23:30

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form. For help clarifying this question so that it can be reopened, see the FAQ.