i have a panel data set. my dependent variable is total costs. and almost all of my independent variables are Categorical variables. like age is "old","new" now i have some questions.
1-should i use a dummy for all of them?(for example, only type variable has 33 values itself) ( or i can use clustering and reduce them?or any other way if you know)
2-is there a difference between behaving Categorical which they have a rank and the ones they don't have?
for example 'type' is "A","B",..."S" so no rank between A and B but quality is "A1","A2","A3" which A1 means highest quality
i dont know why i can find enough information about variable selections and making data ready. so now i have lots of variable and i think i should choose between them and also reduse number of dummies.