Why:include independent qualitative variables to forecasting.
Vs Logistic regression:the dependent (response) Y variable is qualitative.
How: for each value of the qualitative variable X create 1 dummy variable, and assign it 1 when this value appears in a row, otherwise 0. Create total of N-1 dummy variables + constant or N variables without a constant to N values of the variable X.
Example: race: B, W, A. We represent each by D1,D2,D3 respectively. we run regression against Y and these D1,2,3 >>
Y= 34+c2D2+c3D3. Coefficient of D1 is 34 the constant.
Or: Y=c1D1 +c2D2+c3D3
Interaction terms: special values of certain q variables should be studied or paid attention to. Say Black males. So add their combination to regression model as :..+ c2D2E2.
ANOVA: regression using only qualitative vars.
ANCOVA:regression using both qualitative and quantitative vars.
Structural Change test: to test whether an event (happening once) would have any effect on a TS variable. To model this:
A. Add a dummy variable thats 1 for the post-event period and 0 for pre-event. Then check the significance if its coefficient. If the coefficient is high (how high?) then the event did change the variable significantly. This is a better method. Or another method:
B. Use the F-Chow test:
An F test checking if the residuals of the two periods show different behaviour. H0: slopes of the 2regression lines a1=a2 also intercepts b1=b2.