Rosemary Li posted on Tuesday, September 27, 2016 - 6:01 pm
I ran latent growth curve models. Assume I have observations of n firms from year i to year j. For instance, my dependent variable is number of patents. If I understand correctly, the returned intercept means the average number of patents of n firms in year i (initial year).
My question is, if some firms are born after year i (before year j)--observations for these firms before they were born would be treated as missing, how can we interpret the returned intercept? It just feels a bit weird to claim the intercept represents an average initial status of all firms, while some firms haven't born yet in that initial year.
Also, I ran the latent growth poisson model to deal with a count dependent variable. Is there some special model that I should adopt to cope with such an unbalanced panel?
Perhaps you instead want to let time represent the age of the firm instead of year.
I don't understand what you mean by an unbalanced panel.
Rosemary Li posted on Wednesday, September 28, 2016 - 6:23 pm
Thank you for the suggestion, Prof. Muthen. I have thought about it. However, the problem is that I study all firms in the whole industry. So I have firms that are 100 years old as well as firms that are 5 years old. t(age, in this case) seems to be too wide to model (e.g. from 1 to 100). I am new to latent models. If I misunderstood anything, feel free to let me know. Thanks.
By unbalanced panel, I mean not all firms have complete data from year i to year j, due to natural birth and death of firms.