this does sound like fairly easy to package up as a new module,

I found an first-order implementation in hstats (linreg, see below) while inspecting its source code. http://www.sftank.net/code/hstats/src/Math/Statistics.hs But, for some reason, it does not show up in the documentation (not released yet) !? http://hackage.haskell.org/packages/archive/hstats/0.2/doc/html/Math-Sta tistics.html ------------------------------------------------------------------------ -- -- |Least-squares linear regression of /y/ against /x/ for a -- |collection of (/x/, /y/) data, in the form of (/b0/, /b1/, /r/) -- |where the regression is /y/ = /b0/ + /b1/ * /x/ with Pearson -- |coefficient /r/ linreg :: (Floating b) => [(b, b)] -> (b, b, b) linreg xys = let !xs = map fst xys !ys = map snd xys !n = fromIntegral $ length xys !sX = sum xs !sY = sum ys !sXX = sum $ map (^ 2) xs !sXY = sum $ map (uncurry (*)) xys !sYY = sum $ map (^ 2) ys !alpha = (sY - beta * sX) / n !beta = (n * sXY - sX * sY) / (n * sXX - sX * sX) !r = (n * sXY - sX * sY) / (sqrt $ (n * sXX - sX^2) * (n * sYY - sY ^ 2)) in (alpha, beta, r) -----Original Message----- Don, I checked most of them, but did not find anything close. Hstats seems to be the right place, but no, it does not contain such function. It only contains predictive functions in statistics, but there is no best-fit type of functions. I can calculate Sharpe ratio from it, but I can not calculate the alpha and beta between two indexes/funds. http://haskell.org/haskellwiki/Applications_and_libraries/Mathematics FYI -- Under 2.8, HaskellMath links to a 404 page. though, if you're keen.. This may be the only alternative if nothing obvious comes up... Thanks, Steve -----Original Message-----

I am looking for a Haskell module that will do multivariate linear regression. Does someone know which module will do it? That is, the equivalent of Perl's Statistics::Regression.pm.

[1]http://search.cpan.org/~itub/PerlMol-0.35_00.ppm/lib/Statistics/Regre ssion.pm

Thanks, Steve

Always check hackage.haskell.org first, but I'm not sure we have exactly what you're looking for: http://hackage.haskell.org/packages/archive/pkg-list.html#cat:Math in which case the backup is : http://haskell.org/haskellwiki/Applications_and_libraries/Mathematics this does sound like fairly easy to package up as a new module, though, if you're keen.. -- Don ------------------------------------------------------------------------------ Notice: This e-mail message, together with any attachments, contains information of Merck & Co., Inc. (One Merck Drive, Whitehouse Station, New Jersey, USA 08889), and/or its affiliates (which may be known outside the United States as Merck Frosst, Merck Sharp & Dohme or MSD and in Japan, as Banyu - direct contact information for affiliates is available at http://www.merck.com/contact/contacts.html) that may be confidential, proprietary copyrighted and/or legally privileged. It is intended solely for the use of the individual or entity named on this message. If you are not the intended recipient, and have received this message in error, please notify us immediately by reply e-mail and then delete it from your system. ------------------------------------------------------------------------------

Thanks. I have been wondering if there is a Haskell interface to Octave and maybe to Scilab. Steve -----Original Message----- From: Henning Thielemann [mailto:lemming@henning-thielemann.de]

I am looking for a Haskell module that will do multivariate linear regression. Does someone know which module will do it? That is, the equivalent of Perl's Statistics::Regression.pm.

http://search.cpan.org/~itub/PerlMol-0.35_00.ppm/lib/Statistics/Regressi on.pm Maybe you can solve this with the GSL Haskell wrapper and least squares solutions of simultaneous linear equations. (Numeric.LinearAlgebra.Algorithms.linearSolve) http://alberrto.googlepages.com/gslhaskell ------------------------------------------------------------------------------ Notice: This e-mail message, together with any attachments, contains information of Merck & Co., Inc. (One Merck Drive, Whitehouse Station, New Jersey, USA 08889), and/or its affiliates (which may be known outside the United States as Merck Frosst, Merck Sharp & Dohme or MSD and in Japan, as Banyu - direct contact information for affiliates is available at http://www.merck.com/contact/contacts.html) that may be confidential, proprietary copyrighted and/or legally privileged. It is intended solely for the use of the individual or entity named on this message. If you are not the intended recipient, and have received this message in error, please notify us immediately by reply e-mail and then delete it from your system. ------------------------------------------------------------------------------