Clustering functions

group_firms(y,fids; nq=20,ng=10,nstart=100)

groups firms based on empirical cdf pass a vector of firm ids and a vector of wages, at the individual level nq is number of points for the empirical cdf ng is the number of groups nstart is the number of kmeans to run maxiter for each kmeans is set at 1000