It was my fault, as you have seen. Y matrix must contain successes and failures, and not successes and trials. Sorry for my ignorance and thanks a lot for your help. Oscar -------------- next part -------------- An HTML attachment was scrubbed... URL: https://stat.ethz.ch/pipermail/r-help/attachments/20001220/341c2519/attachment.html
On Wed, Dec 20, 2000 at 05:01:44PM +0100, Oscar wrote:> It was my fault, as you have seen. Y matrix must contain successes and failures, and not successes and trials. > Sorry for my ignorance and thanks a lot for your help. > > OscarThis is a very good argument for a more sophisticated modeling paradigm where the lefthand side must be a Binomial object. Filling this up correctly is often easier and more straight forward than trying to figure out what is wanted at fit time. -- +---------------------------------------------------------------------------+ | Robert Gentleman phone : (617) 632-5250 | | Associate Professor fax: (617) 632-2444 | | Department of Biostatistics office: not yet | | Harvard School of Public Health email: rgentlem at jimmy.dfci.harvard.edu | +---------------------------------------------------------------------------+ -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Oscar, PLEASE re-configure your E-mail program, the mail that you sent to R-help contains TONS AND TONS of HTML-"garbage". Please send E-mail to mailing lists __WITHOUT ANY HTML TAGS__.>>>>> "Oscar" == Oscar <orp at intras.es> writes:Oscar> It was my fault, as you have seen. Y matrix must contain successes and failures, and not successes and trials. Oscar> Sorry for my ignorance and thanks a lot for your help. Oscar> Oscar<html> Oscar> <head> Oscar> <meta name="GENERATOR" content="IncrediMail 1.0"> Oscar> </head> Oscar> <BODY background="" bgColor=#ffffff style="BACKGROUND-POSITION: 0px 0px; FONT-FAMILY: Arial" scroll=yes ORGYPOS="0"> Oscar> <TABLE border=0 cellPadding=2 cellSpacing=0 id=INCREDIMAINTABLE width="100%"> Oscar> <TBODY> Oscar> <TR> Oscar> <TD id=INCREDIHEADER width="100%"></TD> Oscar> </TR> Oscar> <TR> Oscar> <TD id=INCREDITEXTREGION width="100%"> Oscar> <DIV><FONT face="Courier New" size=2>It was my fault, as you have seen. Y Oscar> matrix must contain successes and failures, and not successes and Oscar> trials.</FONT></DIV> Oscar> <DIV><FONT face="Courier New" size=2>Sorry for my ignorance and Oscar> thanks a lot for your help.</FONT></DIV> Oscar> <DIV> </DIV> Oscar> <DIV><FONT face="Courier New" size=2>Oscar</FONT></DIV> Oscar> <DIV> </DIV> Oscar> <DIV> </DIV></TD> Oscar> </TR> Oscar> <TR> Oscar> <TD id=INCREDIFOOTER width="100%"> Oscar> <TABLE cellPadding=0 cellSpacing=0 width="100%"> Oscar> <TBODY> Oscar> <TR> Oscar> <TD width="100%"></TD> Oscar> <TD align=middle id=INCREDISOUND vAlign=bottom></TD> Oscar> <TD align=middle id=INCREDIANIM vAlign=bottom></TD> Oscar> </TR> Oscar> </TBODY> Oscar> </TABLE> Oscar> </TD> Oscar> </TR> Oscar> </TBODY> Oscar> </TABLE><SPAN Oscar> id=IncrediStamp> Oscar> <P align=left><FONT face="Arial, Helvetica, sans-serif" Oscar> size=2>__________________________________________________<BR><FONT Oscar> face="Comic Sans MS" size=2><I>IncrediMail</I> - <B>Email has finally Oscar> evolved</B> - </FONT><A href="http://www.incredimail.com/imstamp26.html"><FONT Oscar> face="Times New Roman" size=3><B><U>Click Oscar> Here</U></B></FONT></A></P></SPAN></FONT> Oscar> </BODY> Oscar> </html> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Just as a matter of interest, how did this strange form for the LHS of binomial glms evolve? When my students ask I usually mutter something about a smooth generalization to the multinomial case, but I've no idea if this was really the motivation for doing things this way. Murray Jorgensen At 11:42 AM 20-12-00 -0500, Robert Gentleman wrote:>On Wed, Dec 20, 2000 at 05:01:44PM +0100, Oscar wrote: >> It was my fault, as you have seen. Y matrix must contain successes andfailures, and not successes and trials.>> Sorry for my ignorance and thanks a lot for your help. >> >> Oscar > > This is a very good argument for a more sophisticated modeling >paradigm where the lefthand side must be a Binomial object. Filling >this up correctly is often easier and more straight forward than >trying to figure out what is wanted at fit time. > >-- >+---------------------------------------------------------------------------+ >| Robert Gentleman phone : (617) 632-5250 | >| Associate Professor fax: (617) 632-2444 | >| Department of Biostatistics office: not yet | >| Harvard School of Public Health email: rgentlem at jimmy.dfci.harvard.edu | >+---------------------------------------------------------------------------+ >-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.->r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html >Send "info", "help", or "[un]subscribe" >(in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch >_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._>Murray Jorgensen, Department of Statistics, University of Waikato, Hamilton, New Zealand. [maj at waikato.ac.nz] http://www.stats.waikato.ac.nz/Staff/maj.html_ "These data do not support the hypothesis" Well . . . The first one does, but the second and third don't, now the fourth . . . [ In which I pose the question: is "data" really plural. ] -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._