Professional Documents
Culture Documents
ANSWER
TO temesgen.worku@aau.edu.et
DEADLINE: BEFORE AUGUST 8, 2020 5:00 PM (11:00 LOCAL TIME)
Hypothesis testing in linear regression can be carried out if it can be assumed that the
random error term, is normally and independently distributed with a mean of zero and
variance of t-Test.
The tests are used to conduct hypothesis tests on the regression coefficients obtained in
simple linear regression. A statistics based on the distribution is used to test the two-side
hypothesis that the true slope equals some constant value.
We wish to test hypothesis whether the slope equals a constant β 1,0
H0 : β1 = β1,0
H1: β1 ≠ β1,0
4. Do you think that if heteroscedasticity is present, the conventional t and F tests are
invalid? Why?
Yes; If heteroscedasticity is present, the conventional t and F tests are invalid.
Because, In the presence of heteroscedasticity, the variance of OLS estimator changes
and is larger than in the absence of heteroscedasticity implies the conventional t and F
test measures will be insignificant and do not give valid results.
The t and F test measures vary inversely with variance hence if variance increase due to
heteroscedasticity then the t or F values will be smaller than usual and give insignificant
conclusions.
5. Do you think that in the presence of heteroscedasticity the usual OLS method always
overestimates the standard errors of estimators? Why?
NO;
Because; if heteroscedasticity exist in the regression model then the variance of the estimator will
be an overestimation of the variance in its absence as it depends on the relationship between the
variance and the explanatory variable.
The bias can be positive or negative. Hence, it cannot to asserted that it is always and
overestimation of the standard error of estimators.