Econometrics, Lecture 5, 2014-03-18

1. Goodness of fit:

1. R-squared
2. Adjusted R-squared
3. Information criteria: AIC, BIC

2. F test for regression significance
3. Student’s t test for significance of single parameter

4. Generalized version of Student’s t test

1.

Example 1. Linear Regression Model for sales of oranges: are oranges
and apples substitutes?

2.

Example 2. Security Characteristic Line for Getin Holding: is Getin
Holding an aggressive stock?

5. Generalized version of F test

1.

Example 3. Multiple regression model for advertisement – sales
relation: do the newspaper advertisement expenditures influence the
sales?

Econometrics, Lecture 5, 2014-03-18

Security Characteristic Line (Sharp’s model)

Measuring stock’s risk with β

Security

Characteristic Line

Econometrics, Lecture 5, 2014-03-18

The model

was estimated using n = 24 monthly observations from certain company.
The variables are monthly sales (Y, in millions of PLN), TV advertisement
expenditures (X

1

, in thousands of PLN), and newspapers advertisement

expenditures (X

2

, in thousands of PLN).

The vector of parameter estimates is [3.657 42.23 11.65 -0.6222].
Standard errors of parameters estimates are equal respectively: 23.17;
1.709; 2.250; 2.475.
 
At the 1% significance level verify the hypothesis that the newspapers
advertisement expenditures do influence the sales. The SSR (Sum of
Squared Residuals) in the (unrestricted) model above equals 521.067. SSR
in the (restricted) model with the newspapers advertisement expenditures
excluded equals 3237.74.

t

t

t

t

t

X

X

X

Y



























2

,

2

3

1

,

2

2

1

,

1

1

0