Reading 12.l Qualitative Models & Reading 12.m Economic Meaning

LOS 12.l reads:
“Discuss models with qualitative dependent variables”

It is important to note that to date we have dealt with models that produce QUANTITATIVE dependent variables (e.g. Y). 

An ordinary regression model (e.g everything up to now) is not able to deal with QUALITATIVE dependent variables. 

There are 2 types of models that can deal with qualitative dependent variables

1. Probit & Logits:
 - Probit – Based on the normal distribution
 -  Logit – Based on logistic distribution

These models estimate the likelihood that an event will occur

2. Discriminant models
Similar to Probit & Logit models, but make different assumptions regarding the independent variables.
The model produces a score that produces a Yes or No answer.


LOS 12.m reads:
“ Interpret the economic meaning of the results of the multiple regression analysis and critique the regression model & its results”

This is easy, but important in practice. 

Just because an item is statistically significant does not mean it is economically significant. For example, you might calculate that the share prices rise by 2% in January. If your trading costs are however 3%, it would not make economic sense to buy the shares.