Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

1. The government of Jamaica would like to estimate a model that explains the de

ID: 3221614 • Letter: 1

Question

1. The government of Jamaica would like to estimate a model that explains the

determinants of imports. The following model with its associated variables is proposed by the

Minstry of Finance.

Where

Mt=Imports

Yt=Income

Cg=Government expenditure

Ratiot=ratio of foreign prices to domestic prices

Boomt=Dummy variable representing the boom years from 1974 to 1981

Model 8: OLS, using observations 1967-1991 (T=25)

Dependent variable:m

                                                Coefficient                          std. error                             t-ratio                   p-value

Const                                    -759.871                               1232.98                                 -0.6163                 0.5447

y                                              0.405342                              0.114016                              3.555                     0.0020 ***

cg                                            -0.173065                             0.552459                              -0.3133                 0.7573                  

ratio                                       -891.753                               579.847                                 -1.538                    0.1397  

boom                                    43.9133                                 225.114                                 0.1951                   0.8473  

Mean dependent var                     3556.082              S.D. dependent var                         1531.723             

Sum squared resid                           3198563                S.E. of regression                             399.9102             

R-squared                                           0.943195              Adjusted R-squared                        0.931835             

F(4, 20)                                                 83.02109              P-value (F)                                          3.65e-12                              

Log-liklihood                                      -182.4652             Akaike criterion                                 374.9303                             

Schwarz criterion                              381.0247              Hannan-Quinn                                  376.6207                             

rho                                                         0.129536              Durbin-Watson                                 1.728300                             

Excluding the constant, p-value was highest for variable 12 (boom)

a. Name four assumptions that guide the estimation of a OLS regression model.

b. Use the results of the estimated regression model to answer the following questions:

i. Write down the OLS regression estimated equation.    

ii. Is the overall regression model statistically significant? Prove your answer.

iii. Which variables of the model are statistically significant?    

iv. What is the value of the R2. What is the Interpretation of the value of this R2?  

Alt = 30 + 31Yt + 32cg + 33/. at iot + 34Boomt + Et

Explanation / Answer

a) Assumptions for OLS regression are as follows:

Linearity - the dependent variable is linearly related to the independent variable

Independence - The Yi values are independent of each other

Normality - The residual values hosuld be normally distributed with a mean of 0.

Homoscedasticity - The variance of the dependent variable should not vary with the levels of the independent variable. The residuals should have constant variance.

b) 1) OLS regression estimated quation is as follows:

Mt = -759.871 + 0.405342 Yt - 0.173065 cg - 891.753 ratiot + 43.9133 boomt

2) Yes, the overall regression model is statistically significant as the P value associated with F statistic is very small.

3) The variables for which p value is less that the usual cut off of 0.05 can be termed as statsitically significant. In this case Yt i.e. Income variable

4) The value of R2 is 0.943195. R2 value tells us the percentage of variance explained by the model.