Use the following gretl output for the questions that follow sing observations l

Question

Use the following gretl output for the questions that follow sing observations l 5190 Model 1 Dependent variable HSCORE coefficient std. error t-ratio p-value 465 9.71e-014 0.0624687 0.466337 const 7.94e-091 0.0220056 20 453669 ILLNESS 055 4.58 00967050 15.82 0.152952 ACT DAYS 0.1956 0.0216199 1.295 0.0279870 PRESCRIB 0.0008 3.370 0.038 4299 0.129508 NONPRESC 0.0205 2.318 0.172298 0.0743441 INCOME 2.1242 66 dependent var Mean dependent var 1.217534 Sum squared resid 19313.60 s.E. of regression 1.930186 0.174380 Adjusted R-squared R-squared 0.175176 1.2e-213 P-value (F) F (5, 5184) 220.1949 Akaike criterion 21560.63 Log-likelihood 10774.32 21574.39 Schwarz criterion 21599.96 Hannan-Quinn Where HSSCORE is a score for overall health of person i Where ILLNESS is the number is illnesses person I has had in the past two weeks of person i Where ACTDAYS is the number of days of reduced activity in the past two weeks of person i Where PRESCRIB is the total number of prescribed medications used in the past two days by person i Where NONPRESC is the total number of nonprescribed medications person I used in the past two days. Where INCOME is the annual income (in thousands of currentAustralian dollars) of person i Correlation coefficients, using the observations 1-5190 5% critical value (two-tailed) 0.0272 for n 5190 INCOME ILLNESS ACTDAYS PRESCRIB NON PRESC 1.0000 -0.1488 -0.0475 -0.1950 0.0378 INCOME 1.0000 0.2181 0.4226 0.1570 ILLNESS 1.0000 0.2514 0.0393 ACTDAYS 1.0000 -0.0435 PRESCRIB 1.0000 NONPRESC White's test for heteroskedasticity (squares only) OLS, using observations 1-5190 Dependent variable: uhat coefficient std. error t-ratio p-value const 1.92713 0.509001 0.0002 ILLNESS 1.57989 0.306168 5.160 0.214180 1.425 2.56e-07 ACTDAYS 0.305305 0.1541

Explanation / Answer

a) No multicollinearty

because

1:In practice, usually a VIF > 5 or 10 indicates that the associated regression coefficients are poorly estimated because of multicollinearity.

2.Off diagonal element is not close to one.

b)according to White test P<alpha reject H0.the model is heteroscedastic.

Related Questions

Navigate

• Browse (All)
• Browse U
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.