The class forecasting exercise results

1
The class forecasting exercise results
We want to take you through the results,
interpret the importantstatistics, and perform
the forecastof domestic car sales.
2
(No Transcript)
3
(No Transcript)
4
Model 1
5
(No Transcript)
6
Standardized Regression Coefficient (Beta)
Example Beta for DPI, Model 1
Meaning, a one standard deviation changein DPI
will result in a change in DCSequal to
7
The regression equationobtained from the model1
specification is givenby the following
8
Interpretation of the estimates

• A 235 increase in per capita disposable income
  added an average of 1 unit to quarterly car sales
  during the sample period, ceteris paribus.
• A 1 percentage point decrease in the prime rate
  added an average of 12.34 units to quarterly car
  sales during the sample period, controlling for
  changes in DPI.

9
DCSt ?0 ?1DPIMA ?2PRt ?3Moodt
Model 2
10
(No Transcript)
11
Model 3
DCSt ?0 ?1DPIMA ?2PRt ?3Moodt
Seasont
12
(No Transcript)
13
The F test

• The F test provides another goodness of fit
  criterion for our regression equation. The F test
  is a test of joint significance of the estimated
  regression coefficients.
• The F statistic can be computed as follows

• where K - 1 is degrees of freedom in the
  numerator and n K-1 degrees of freedom in the
  denominator.
• K is the number of explanatory variables.

14
?We set up the following null hypothesis (H0) and
alternative hypothesis (HA)H0 ?1 ?2 ?3
?4 0 HA H0 is not true.
?We adhere to the following decision rule Reject
H0 if F gt FC, where FC is the critical value of
F (obtained from F table).

• In our case, F 29.094
• FC with three (3) df in the numerator and
  forty-nine(49) df in the denominator is equal to
  (approximately) 2.84.
• Reject H0 since 29.094 gt 2.84

15
(No Transcript)
16
In-sample forecast for 98.1 to 00.4
17
Our forecast