| Ticket UUID: | 1838731 | |||
| Title: | Pretty quick inverse normal cdf | |||
| Type: | Bug | Version: | None | |
| Submitter: | erickb | Created on: | 2007-11-26 14:17:25 | |
| Subsystem: | math :: statistics | Assigned To: | arjenmarkus | |
| Priority: | 5 Medium | Severity: | ||
| Status: | Closed | Last Modified: | 2009-11-16 15:19:56 | |
| Resolution: | Fixed | Closed By: | arjenmarkus | |
| Closed on: | 2009-11-16 08:19:56 | |||
| Description: |
Hello, I have written a pretty quick inverse normal cdf proc. I start with a decent (I think good to 10^-4 ??) approximation from Abramowitz and Stegun as a starting point and then use Newton's method to find a much better estimate. Use it as you like. I'm attaching the code, and providing it as public domain (no copyright). Incidentally, math::statistics::pnorm is much faster than math::statistics::cdf-normal, and it might be better to use it (??) Eric | |||
| User Comments: |
arjenmarkus added on 2009-11-16 15:19:56:
allow_comments - 1 Closing this issue - incorporated since january 2008 erickb added on 2007-11-26 21:17:25: File Added - 256104: norminv.tcl | |||
Attachments:
- norminv.tcl [download] added by erickb on 2007-11-26 21:17:25. [details]