I have a question about a conditional logit model that I want to use to see how characteristics of various possible destination countries affect migration decisions.
I have data on migration flows between 14 countries, as well as various characteristics of these countries. I calculated that there are 14*13=182 cases in my model (14 countries of origin with 13 possible destinations). I listed them, so that every line in my dataset represents a combination of a country of origin and a country of destination.
My problem now is: I can’t seem to wrap my head around what now is my dependent variable. In the examples of conditional logit models I found so far, the dv is always a variable representing the choice made by individuals, with values ‘1’ and ‘0’. However, in my case in fact all lines represent the choice for a destination, since there are observations for every origin-destination combination. In addition, one line in my dataset does not represent one person, but a flow of hundreds of migrants. I expect I have to do something with frequency weights here, but I do not understand what exactly.
I would really appreciate some advice on these issues! Many thanks in advance!