inv_normal is a more accurate implementation of NORMSINV. Use Harlan's
formula with either.

Jerry

Paul D. Simon wrote:

Thanks very much for your response, Jerry.

I tried the inv_normal function, but I couldn't figure out how to apply
it properly.

For example, the number of Standard Deviations for 68.3% is 1; for 95%,
it's 1.95996. When applying the inv_normal function to 68.3%, it gave
me an answer of 0.476104403489395. For 95%, it gave me an answer of
1.64485362695147. Obviously, I doing something wrong.