An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20090209/1457e71e/attachment-0001.pl>
Calculating variables
2 messages · Jesús Guillermo Andrade, Uwe Ligges
Jes?s Guillermo Andrade wrote:
Dear fellows: This is the problem: I have 5 variables A, B, C, D and E with a range from 1 to 100 with 0.1 steps. Depending on the different values these have, the results of the formula change: alitemp <- ((Abase/llmcc$Clase)*PClase)+(((1/llmcc $Categoria)*Abase)*PCategoria)+((Abase*llmcc$Phi)*PPhi)+((Abase*llmcc $Rf)*PRf)
There is no instance of A...E in that formula, is it?
So, alitemp (a 283 element series) changes if any of the P** values changes. I need to find a combinations of those 5 variables that approximates the desired results. Although I can make use of a 'for' loop, Im not sure of how to combine the whole 5 variables changing at once. I understand there are many possible combinations (factorial of 5000!, 422 with some 500 digits) but I surmise there must be a better way to do it. So far, I have been changing them manually, but this task make the whole thing terribly slow and inneficient. Any ideas are appreciated. Thanks.
What do you really want to know? Some prediction for anything? Or just the time t when 5 variables are changing between t-1 and t? If the latter, calculate first order differences and check for t where all 5 first order differences are not equal to zero. Uwe Ligges
Cuanto mas grande son, mas da?o hacen. Postulado de Perkin. ------------------------------ Jes?s Guillermo Andrade (Abg.) Gerente de Litigios y Corporativo. EDM. AC. API. Andrade & Moreno S.C. (http://amlegal.wordpress.com/) [[alternative HTML version deleted]] ------------------------------------------------------------------------
______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.