stone@cunixb.cc.columbia.edu (Glenn Stone) (12/19/89)
Can anyone recommend software for conducting linear programming
analysis on a 286-based machine? What I'd really like is an add-in
for Lotus 123, but a stand-alone would work too.
Note that linear programming is not a way of writing source code,
but a multivariate statistical technique.
Also any suggestions for other newsgroups for this posting would
be appreciated.
Thanks --
G. D. Stone
Dept. of Anthropology
Columbia Univ.LC.YRS@forsythe.stanford.edu (Richard Stanton) (12/19/89)
In article <2516@cunixc.cc.columbia.edu>, stone@cunixb.cc.columbia.edu (Glenn Stone) writes: >Can anyone recommend software for conducting linear programming >analysis on a 286-based machine? What I'd really like is an add-in >for Lotus 123, but a stand-alone would work too. > There is a linear programming add in for LOTUS 1-2-3 called "What's Best". I don't know who produces it, but I THINK they are related to the people who produce LINDO (a standard stand-alone linear programming package for the PC / many other machines). Richard Stanton pstanton@gsb-what.stanford.edu
ts@uwasa.fi (Timo Salmi LASK) (12/19/89)
In article <2516@cunixc.cc.columbia.edu> stone@cunixb.cc.columbia.edu (Glenn Stone) writes: >Can anyone recommend software for conducting linear programming >analysis on a 286-based machine? What I'd really like is an add-in >for Lotus 123, but a stand-alone would work too. I don't know about the recommending bit, but if you have the (anonymous) ftp capability you can get /pc/ts/tslin29.arc from our site. This file contains a linear programming and a linear goal programming program. >Note that linear programming is not a way of writing source code, >but a multivariate statistical technique. Sorry, but you have confused things in here. Linear programming is NOT a multivariate statistical technique (even if it can be applied as a tool in least absolute deviations method in multiple regression analysis.) Linear programming is a method for maximizing (or minimizing) a linear function under linear constraints. The traditional LP task in vector notation is max z = c'x subject to Ax < b - x > 0 - If you are looking for a newsgroup for LP questions, I think that your best bet is sci.math.num-analysis. ................................................................... Prof. Timo Salmi (Site 128.214.12.3) School of Business Studies, University of Vaasa, SF-65101, Finland Internet: ts@chyde.uwasa.fi Funet: vakk::salmi Bitnet: salmi@finfun
stanwass@uxh.cso.uiuc.edu (12/21/89)
Linear programming is not a statistical technique. Stan Wasserman Psychology and Statistics University of Illinois
kaleb@mars.jpl.nasa.gov (Kaleb Keithley) (12/22/89)
In article <2516@cunixc.cc.columbia.edu> stone@cunixb.cc.columbia.edu (Glenn Stone) writes: >Can anyone recommend software for conducting linear programming >analysis on a 286-based machine? What I'd really like is an add-in >for Lotus 123, but a stand-alone would work too. Cal State University, Northridge has a product in their Business Dept. computer labs called STORM. It uses a spreadsheet style interface, with some prompting for non-calculation type parameters. It does linear programming and a lot more (transportation problems, PERT/CPM, etc). I don't know if this was developed in house or if it's a commercial product, I used it last semester, and while the user interface is less than intuitive, its' performance on a 5Mhz PC was acceptable. Chewey, get us outta here! kaleb@mars.jpl.nasa.gov (818)354-8771 Kaleb Keithley