Hello all,

I have 3 years of cross sectional data of which 2 years have information on my independent variable (weight), however my dependent variable (long term illness) only appears in the 3rd wave. I will like to know if it is possible to solve for ommited variable bias using individual fixed effects and is it also possible to use lagged dependent variable as an instrument for instrumental variable analysis. i also found this paper (https://www.nber.org/papers/w11796.pdf) on birth weight but i do not understand how they came about using fixed effects sincebirth weight is only measured onceimplying that it is not a longitudinal data. I will really need answers to these questions. Many thanks.