Dummy_Variable and and Selection in Factor Analysis


#1

Hello,
I am new to the Data science and I was working on a dummy model for linear regression. There were more than 53 variable which are numeric in nature but data Data Dictionary stating these are Cat var.
E.g. agecat
1 <18
2 18-24
3 25-34
4 35-49
5 50-64
6 >65
9 No response

So I have created dummy Variables for the those var.

Now questions are.

  1. Should I create dummy Var for all the variables?
  2. If there is any missing Value in this Variable , what should be imputed ?
  3. Which Variable must be given in Factor Analysis Dummy Variables or original Variables?

#2

HI @purushottamnarayansh

This is one ordinal variable ( cat 1 < cat 2 ) with exception of no response, base on the problem you solve this encoding could be better. You can encode in order and give 0 = no response (but this will have a weight) to non response and start with one for < 18.

Hope this help
Best regards
Alain