Conditional Heteroskedasticity

I was doing a problem and came across this in the answer explanation: “Conditional heteroskedasticity is a case in which the error variance is related to the magnitudes of the independent variables (the error variance is “conditional” on the independent variables). The consequence of conditional heteroskedasticity is that the standard errors will be too low, which, in turn, causes the t-statistics to be too high.” I thought the standard errors may be too high OR too low with conditional hetero. So t-stats will be misrepresented, but they may be too high or too low.

Dude I think that as the number of your variables increase, the coefficient of determination increases, which implies RSS increases and SSE remains the same. So, in that way, as the magnitude of variables increase your SEE will become low and would cause the T-stat to be high causing Type 1 errors… So the standard error is always low in the scenario you specified…

CFAI pg. 305 “…the most likely result of heterskedasticity is that the estimated standard errors will be underestimated and the t-statistics will be inflated.” So ok I get it. It’s the MOST likely result. But low standard errors is not the necessary result all the time. According to Schweser pg. 197, standard errors and t-stats may be higher or lower. Whatever. I’m pretty sure I understand what’s going on. Let’s put it to rest.