When models exhibit heteroscedasticity, least squares is the most ...

[lbcchick]

When models exhibit heteroscedasticity, least squares is the most efficient procedure for estimating the coefficients of the regression model.
 
  Indicate whether the statement is true or false

Question 2

On the basis of a random sample the null hypothesis: H0 :   0 is tested against the alternative: H1 :  > 0 and the null hypothesis is not rejected at the 10 significance level.
 
  Does this necessarily imply that 0 is contained in the 90 confidence interval for ?

[mk6555]

Answer to Question 1

FALSE

Answer to Question 2

No, the 90 confidence level provides for 5 of the area in either tail. This does not correspond to a one-tailed hypothesis test with  = 10 which has 10 of the area in one of the tails.

[lbcchick]

Gracias!

[Zebsrer]

Thanks for the timely response, appreciate it