1
Enter
data into the spreadsheet following normal WinORS conventions; that is, several
independent variables and at least one dependent variable.
2
Enter
names on row one and one-letter
variable types (VType) across row two (I, D, F,
G, or W).
3
Choose
Menu path: Applications / Statistical Methods / Regression / Ordinary Least
Squares.
If the data is not saved to disk WinORS will force you to File/Save at
this point. If the block range of the
data is not already set, then use the roll-up arrow and do so at this time.
4
Proceed
to generate a solution. Analyze the
regression results. First check the
validity of the regression parameters.
When multicollinearity is present (see: Variance
Inflation Factors (VIFs) drop or
transform offending variables (see Multicollinearity
- Solving the Problem ).
Deselect (remove the I variable types) collinear variables.
5
Return
to step #3 and solve the revised model except in this case choose
OLS Optimze settings. OLS
Optimize automatically creates and employs either the Durbin adjusted or
First-differenced data as a means of eliminating serial correlation from the
data set under analysis. WinORS reads
the Durbin-Watson statistic at the 95% confidence level (2-tailed test) to
determine whether the correction is necessary.
After solving for the OLS parameters on the time-series adjusted data,
WinORS will attempt to correct for heteroskedasticity based on the p-value of
White's test. If the p-value of the
White's test is greater than 0.05 a weighted OLS analysis is performed. Upon the completion of all steps, the WinORS
user is presented with the corrected solution across all relevant tabs.
a.
Durbin
Adj -- check this box
is you want auto-correlation correction using the Durbin adjustment method.
b.
First
diff. --
check this box if you want auto-correlation correction using first
differenced data.
c.
WOLS
-- check this box if you want to correct for heteroscedastic variance
after correcting for auto-correlation violations.