| home | help | back | first | fref | pref | prev | next | nref | lref | last | post |
Date: 27 Feb 95 13:29:25 EST From: "Steven P. Millard" <74273.3130@compuserve.com> To: S-news <S-news@utstat.toronto.edu> Hello, I've got a client who is looking for a DOS- and/or WINDOWS-based package with the capability of performing multiple regression. He wants to incorporate the multiple regression part into software he is developing, so he'd like a package that has an up-front fee with license to freely distribute any code from it that you include in your own package. He has a copy of something called ProMath 2.0 from a company called Tera Tech, which is a set of macros (written in Fortran) to do mathematical and simple statistical computations. (I think he said that it costs about $100). This would be exactly what he wants, except that it only performs simple regression (although it does do matrix manipulation, so it wouldn't be too hard to program multiple regression using the (numerically naive?) method of inv(X'X)X'Y). One person suggested MicroSoft IMSL libraries and also the NAG libraries. The problem with using these is that it's really overkill (price-wise) for what my client wants. MS Fortran 5.1 is about $170 and the corresponding IMSL libraries are about $470. Does anyone have any other suggestions? Thanks for your time! --Steve M. (~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~) ( _____________ ) ( | *** | Steven P. Millard ) ( | * | ) ( | * * * | P robability, ) ( | * * * | S tatistics & ) ( | * | I nformation ) ( | * | ) ( | *** | 7723 44th Avenue NE ) ( |___________| Seattle, WA 98115 USA ) ( ) ( Voice/FAX: (206) 528 - 4877 ) ( E-mail: 74273,3130@CompuServe.COM ) (___________________________________________________________)
| home | help | back | first | fref | pref | prev | next | nref | lref | last | post |