Is there a way to calculate the optimal sample size for a balanced panel over 8 years from a population of approximately 30,000 and involving 10 variables being studied, say for 95% degree of confidence?
3
-
2$\begingroup$ I think you're going to need more information to do this. EG, how often will the units be measured, once per year? What parameter values do you expect to find / will you want to be able to differentiate from 0, etc. $\endgroup$– gung - Reinstate MonicaCommented Nov 25, 2012 at 19:12
-
$\begingroup$ You need to define the optimality criteria. Is it based on precision, cost or something else. $\endgroup$– djhurioCommented Nov 25, 2012 at 19:52
-
$\begingroup$ thanks for your replies, i am seeking to measure my units once a year for 8 years . . . my aim is to find out why the price charged to the customer varies from one year to another (e.g. as the value of the client's assets increases etc... other parameters will involve a 0 or a 1, say as supplier is changed) $\endgroup$– user17160Commented Dec 4, 2012 at 19:48
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
You might also be interested in the answers of these questions.
