SPSS: Comparing regression coefficient from multiple models

问题

Assuming I have price of houses as the dependent variable and the following as the independent variable:

1. Age
2. Area
3. Floor
4. Time taken to walk to nearest railway station (time_walk)
5. Time taken to commute, via train, from the nearest station to the CBD station (time_train)

Is there a way to compare the coefficient of time_walk given different ranges of time_train. In essence what I would like to achieve is to investigate if people value walking differently, given a change in the mrt traveling time.

• Model A: (0-9 mins time_train): How would Walking_Time affects house pricing?

• Model B: (10-19 mins time_train): How would Walking_Time affects house pricing?

• Model C: (20-29 mins time_train): How would Walking_Time affects house pricing?

I understand that I can't create 4 models, each containing only the relevant details (eg. 0-9 mins train time, 10-19 mins train time ...etc) as the n number would be different. This comparing the coefficient estimates wouldn't be fair.

回答1:

What you are suggesting is that the effect of walking_time is dependent on time_train. That's an interaction hypothesis! If time_train is an ordinal level variable, you can add interaction terms for each category dummy with the variable walking_time. If time_train is a continuous variable, one extra term is sufficient (walking_time * time_train).

来源：https://stackoverflow.com/questions/25399461/spss-comparing-regression-coefficient-from-multiple-models