Question

Suppose you conduct the modified White’s Test. You estimate the regression with the squared residuals as the dependent variable and predicted values and predicted values squared as independent variables. From the regression results, you get the F-statistic is 0.353 with a p-value of 0.414. At a Level of Significance of 5%, what does the modified White’s Test say about the Linear Regression?

Group of answer choices

Heteroskedasticity is unlikely to be a problem with the Linear Regression

Heteroskedasticity is likely to be a problem with the Linear Regression

The coefficients of the Linear Regression are likely biased

The Linear Regression doesn’t have a good Goodness-of-Fit

Answer 1

P-value rule for a two-tailed test:--

if the p-value is less than the alpha/2 then reject the null hypothesis

if the p-value is greater than the alpha/2 then fail to reject the null hypothesis

H0: no heteroscedastic

Ha: heteroscedastic

here, p-value = 0.414

alpha/2 = 5%/2 = 0.05/2 = 0.025

0.414 > 0.025

=> p-value > alpha/2

=> fail to reject null hypothesis

hence no heteroscedastic

hence correct option is A i.e. Heteroskedasticity is unlikely to be a problem with the Linear Regression