Suppose from a regression on 74 observations of a dependent variable on a constant and one independent variable, the SSR = 8.916
economics
Description
Suppose from a regression on 74 observations of a dependent variable on a constant and one independent variable, the SSR = 8.916
If the sample is divided into the first 30 observations and the last 24 observations and the
same model is estimated on each subsample to get SSR1 = 1.215 and SSR2 = 3.4895
Carry out a test, to determine if there is heteroscedasticity in the model
What is the Null Hypothesis?
What are the test statistic and its distribution?
Test at the 1% level. What is the critical value?
What is the decision?
2. How does Heteroscedasticity affect the least square estimator and forecasts?
3. Suppose the error variance of a model is describe by the equation
Describe how to use Feasible Generalized Least Squares to get efficient estimates of the
coefficients of the model.
4. When two are more independent variables are correlated with one another the value of the R
2
will be low . True or Falseg
5. You estimate a model with 3 variables. You then add a fourth, the value of R
2
increase by a
small amount, the new estimated coefficient is statistically significant, the other coefficients
change substantially and all their standard errors increase substantially. What is likely taking place?
