I want to generate a range of numbers that follow a Pareto distribution.

I want these numbers to assume a mean value of say 1000.

With the pareto distribution I need to specify my shape and location parameters (or alpha and beta depending on the terminology you would like to use).

In calculating my shape and location parameters, do I simply choose

1< shape <2 and then calculate my location value from

mean = (shape*location)/(location - 1)

or am i misunderstanding the approach ?