Hello,


My aim is to measure labour productivity in firm-level panel data. (all the analysis would be based on aggregate data)
And then, I need to get a graph of x axis: log(labour productivity) and y axis: percentage of firms.

1) First of all, labour productivity here is estimated by: ln(gross value added/employment)
a) Since it is a panel data, I noticed that I have to aggregate gross value added and employment first. (I am not sure whether I need to do this in this step or not...)
But the problem is that I do not know how to aggregate this value. Do I have to use collapse command to compute this ? If yes, do I have to aggregate by mean?
b) Or, do I have to aggregate the value of log(labor productivity) after computing them each year ?

2) In order to get a graph (kdensity), what would be the command to draw <<x axis: log(labour productivity) and y axis: percentage of firms>> ?


In sum, I am not sure when should I aggregate the values and how can I aggregate them..
Sorry if my questions are unclear and thank you in advance.

Anne-Claire