Hello, know somebody a "nice" strategy to analyze a lot of binary variables with hundred to thousands of cases. P.S. One nice example for this and something more is the configurational approach from C.Ragin http://www.nwu.edu/sociology/tools/qca/qca.html ,but i fight with the complexity of my data and the speed of the contibuted software in TCL/TK and would attempt to implement this in R ! thanks for any suggestions christian -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- 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 _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
On Mon, 18 Feb 2002, christian wrote:> Hello, > know somebody a "nice" strategy to analyze a lot of binary variables with hundred to thousands of cases.It depends on the structure of the data, in particular which variables are responses and which are explanatory, as well as what `a lot' means, since the number of cases quoted is small. (There are plenty of examples with millions or more cases and hundreds of variables.) The standard approaches are log-linear models for joint responses, and logistic regression for single ones. There are more sophisticated ones involve selecting graphical models, but these need more input from the subject matter. The data mining community has a number of visualization methods, ....> P.S. > One nice example for this and something more is the configurational approach from C.Ragin > http://www.nwu.edu/sociology/tools/qca/qca.html ,but i fight with the complexity of my data > and the speed of the contibuted software in TCL/TK and would attempt to implement this in R !Does any expert statistician recommend that approach? -- Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272860 (secr) Oxford OX1 3TG, UK Fax: +44 1865 272595 -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- 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 _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
> > >>P.S. >>One nice example for this and something more is the configurational approach from C.Ragin >>http://www.nwu.edu/sociology/tools/qca/qca.html ,but i fight with the complexity of my data >>and the speed of the contibuted software in TCL/TK and would attempt to implement this in R ! >> > >Does any expert statistician recommend that approach? >Sorry only me and the author, but i think it is an nice alternative to see cases as configurations to classic descriptives, but i mean not that classic methods are bad ! easy Example: 4 independent variables and one dependend which have only the state true and false ! men , age > 40, catholic , income > 30000EUR and the depedend variable "elect labor party" ...so you have not 4 independend variables - in this approach you have got 16 configurations which are different to the outcome . Further you define benchmarks for analyze neccessary & sucessfully conditions and test the signficance ! A way further Ragin execute how it is possible to work with fuzzy-sets instead of crisp-sets, too ! regards, christian schulz> >-------------- next part -------------- An HTML attachment was scrubbed... URL: https://stat.ethz.ch/pipermail/r-help/attachments/20020219/727301b1/attachment.html