![fgls eviews fgls eviews](https://slideplayer.com/slide/13124815/79/images/81/Cochrane-Orcutt+Estimator.jpg)
Heteroskedasticity Tests > White (Include Cross Terms). Eviews: Estimate Equation > Output > View > Residual Diagnostics >
![fgls eviews fgls eviews](https://i.ytimg.com/vi/x7m3uhTGcbk/hqdefault.jpg)
May have low power (low probability of correctly rejecting H 0 ) and is non-Ĭonstructive (doesn’t tell us what to do next). Tests for relationships between EVs and the error -> if it has a high Where R 2 is the coefficient of determination regressing ei 2 on a constantĪnd all explanatory variables, their squares and their cross products. H 0 : σi 2 = σ 2 for all i H 1 : not H 0. O Plot the residuals (or squared residuals) Methods for Detecting Heteroskedasticity If there is a heteroskedasticity problem, the BLUE is the Generalised Least Squares (GLS) model.
![fgls eviews fgls eviews](https://image.slidesharecdn.com/1-150614024536-lva1-app6891/95/m-hnh-hi-quy-tuyn-tnh-vi-d-liu-cho-1-638.jpg)
O OLS is inefficient (no longer BLUE: Best Linear Unbiased Estimator = smallest variance). O Standard errors are no longer given by standard formulas -> effects HTs and CIs. Consistent: the variance gets smaller as the sample size gets larger. Unbiased: the estimations are right, on average, even when the sample sizes are Red arrows = the variance is increasing over time -> Heteroskedasticity Heteroskedasticity is more commonly observed in cross-sectional data, rather than time-series O Allows each observation to have it's own individual (hetero = different) variance. O Basically says that all individual observations share a common (homo) variance (spread).