BookmarkSubscribeRSS Feed
☑ This topic is solved. Need further help from the community? Please sign in and ask a new question.
Lapis Lazuli | Level 10

## Is there a SAS equivalent of qnorm() for non-standard normal distributions?

How does one in SAS 9.4 generate the 20th percentile of a normal distribution with, say, a mean of 500 and standard deviation of 50? In R I call: qnorm(.20, 500, 50)

I have a bunch of percentiles of reading achievement scores, and I need to convert them to to scale scores.

I'm looking throughout the documentation and can't seem to find a way...unless I need to use PROC IML first, but I'm not very good with that.

1 ACCEPTED SOLUTION

Accepted Solutions
Super User

## Re: Is there a SAS equivalent of qnorm() for non-standard normal distributions?

Does the value of X look like what you want:

```data junk;
x = quantile('normal',.2,500,50);
run;```

The SAS Quantile function takes as the first parameter the name of the distribution, there are about 25. Then the quantile of interest and then distribution parameters. If you leave off the mean and standard deviation the distribution for Normal will assume 0 and 1.

3 REPLIES 3
Super User

## Re: Is there a SAS equivalent of qnorm() for non-standard normal distributions?

Does the value of X look like what you want:

```data junk;
x = quantile('normal',.2,500,50);
run;```

The SAS Quantile function takes as the first parameter the name of the distribution, there are about 25. Then the quantile of interest and then distribution parameters. If you leave off the mean and standard deviation the distribution for Normal will assume 0 and 1.

Lapis Lazuli | Level 10

## Re: Is there a SAS equivalent of qnorm() for non-standard normal distributions?

Nice! I read about that function, but did not get that I could implement it that way.
Super User

## Re: Is there a SAS equivalent of qnorm() for non-standard normal distributions?

@svh wrote:
Nice! I read about that function, but did not get that I could implement it that way.

Understand. Most of those other distributions really want one or more parameters and the examples using Normal in that function don't show the optional mean and std dev.

Discussion stats
• 3 replies
• 252 views
• 2 likes
• 2 in conversation