[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Cluster option in Regress command
Use the -xt- commands for time series-cross sectional
data, not the cluster option.
Dr Sumon Bhaumik wrote:
> Hi everyone,
> I need help with a Stata command. In the "regress" command, there
> is an option "cluster." Apparently, it is used when the data is
> such that some of the observations may be correlated. For example,
> if I estimate a specification using time series-cross section
> pooled data of n banks and t years. Since the portfolios and many
> of the financial ratios of each of the banks, used in the
> specification, may be correlated across the t years, it seems to
> me that I should use the "cluster" option. Am I mistaken? What
> exactly is the econometrics of the "cluster" option? Thanks!
* For searches and help try: