I retract my objection (sort of). I was thinking of the idealized maximum log-likelihood objective function, for which the transformation+linregress approach is technically inadequate. But it turns out that in practice, transforming and then applying linear regression is a close enough approximation that it is indeed used.
Thanks, David -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of David Shin Sent: Tuesday, April 13, 2010 12:44 PM To: 'Peter Johansson' Cc: '[email protected]' Subject: RE: [Help-gsl] logistic regression support That does not make sense to me. Logistic regression applies a hinge function to the linear combination of the inputs, not to the individual inputs themselves. The training of a logistic regression model cannot be framed as a composition of a data transformation followed by a linear regression. -----Original Message----- From: Peter Johansson [mailto:[email protected]] Sent: Tuesday, April 13, 2010 12:42 PM To: David Shin Cc: '[email protected]' Subject: Re: [Help-gsl] logistic regression support David Shin wrote: > Does GSL provide support for training logistic regression models? > > There is support for linear regression http://www.gnu.org/software/gsl/manual/html_node/Linear-regression.html so you can transform your data so the logistic model becomes a linear model and then use the linear regression provided by GSL. Cheers, Peter -- Peter Johansson svndigest maintainer, http://dev.thep.lu.se/svndigest yat maintainer, http://dev.thep.lu.se/yat Note: This email is for the confidential use of the named addressee(s) only and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you are hereby notified that any review, dissemination or copying of this email is strictly prohibited, and to please notify the sender immediately and destroy this email and any attachments. Email transmission cannot be guaranteed to be secure or error-free. Jump Trading, therefore, does not make any guarantees as to the completeness or accuracy of this email or any attachments. This email is for informational purposes only and does not constitute a recommendation, offer, request or solicitation of any kind to buy, sell, subscribe, redeem or perform any type of transaction of a financial product. _______________________________________________ Help-gsl mailing list [email protected] http://lists.gnu.org/mailman/listinfo/help-gsl Note: This email is for the confidential use of the named addressee(s) only and may contain proprietary, confidential or privileged information. If you are not the intended recipient, you are hereby notified that any review, dissemination or copying of this email is strictly prohibited, and to please notify the sender immediately and destroy this email and any attachments. Email transmission cannot be guaranteed to be secure or error-free. Jump Trading, therefore, does not make any guarantees as to the completeness or accuracy of this email or any attachments. This email is for informational purposes only and does not constitute a recommendation, offer, request or solicitation of any kind to buy, sell, subscribe, redeem or perform any type of transaction of a financial product. _______________________________________________ Help-gsl mailing list [email protected] http://lists.gnu.org/mailman/listinfo/help-gsl
