Hi all.

I am working on a research project to estimate the respective effects of divorce, jobloss and death of a family member on household income. I am using a dataset called NIDS, which is South Africa's first nationally representative household panel dataset. There are currently 5 waves of the dataset, which were conducted every two years from 2008-2016. These waves were merged together and a balanced panel of 130 000 observations was retrieved.

My supervisor advised that I use wave 1(2008) as baseline, apply a shock in wave 2(2010) and separately estimate the effect these shocks have on household income in wave 3(2012), 4(2014) and 5(2016). The shocks are all dummy variables. My supervisor suggested a model with lags, but I have not found any information on something similar to such a question.

I would very much appreciate any suggestions as to how one could empirically model such a question in Stata.

Kind regards.