Error term heteroskedasticity in trend models?

CFA Level II Quant
#1

Hi everyone,

Not being an econometrics specialist, I have three questions I hope you can help me with:

  1. Does error term heteroskedasticity create biased coefficients’ standard errors in a linear or log-linear trend model? In the curriculum, they discuss heteroskedasticity focusing on AR, MA and ARMA models, presenting ARCH. Don’t know if it heteroskedasticity -in general, not only related to ARCH- matters in linear and log-linear trend models.

And if error term heteroskedasticity in linear and log-linear trend matters:

  1. Is it OK to use ARCH in a linear or log-linear trend model to evaluate and model error term heteroskedasticity? If not so, why?

  2. It is OK to use the Breusch-Pagan test in a linear or log-linear model to evaluate error term heteroskedasticity? If not so, why?

Thanks for reading this!