You are not logged in. You can browse but not post. Login or Register by clicking 'Login or Register' at the top-right of this page. For more information on Statalist, see the FAQ.
Anam:
welcome to the list.
Strange question indeed.
As far as I know there's no such test.
You may have to deal with large N (N=cross-sectional dimension), small T (T=time-series dimension) panel datasets or the other way round.
Please, provide furter details to help interested listers helping yourself.
thank u carlo for responding thats very very kind of u...... i was doing a panel data analysis by taking 10 years data on 18 cross sections...my course instructor asked why are u taking 18 cross sections??? there are tests which tell how many cross sections can be taken for certain no of years.....go and look for them on internet.... i have been searching for about a week now but failed to get any answer......
Anam:
it also sounds strange that you can choose yourself how many cross section units to include in your panel data analysis (by the way: according to which requirement did you select them? Were they randomly selected?)..
Usually, a panel dataset is composed of a given N and T dimensions.
Anam:
it still sounds strange to me that you can choose yourself how many cross section units to include in your panel data analysis.
Anyway: what's your research goal? Simply creating a panel dataset? A large N, small T panel dataset or the other way round?
thank u so much carlo for responding.... thats very kind of u....... if one can not choose cross section by his self then how are they be chosen???? whats the criteria??? how do people select a pannel??????
my research goal is a cross country analysis of economics of health... i m analyzing impact of income, education and socioeconomic conditions on health status by using infant mortality as a measure of health status........... i was interested in taking large N and small T
Comment