hello all, I am a beginner at R and not exactly a statistician either. I'm messing around with a probit model on an enconomic time series. I can get the model estimated but I have not been able to get it to give me predictions out of sample data. I'm using the predict function but getting errors. Here is my code:
#predict a recession 12 months out #k=12 means 12 months forward #dates must be trimmed manually recession.probModel <- glm(formula = window(lag(recession.TS,k=12),start=c(1986,2)) ~ window(FEDFUNDS.TS,start=c(1986,2),end=c(2010,7)) + window(sp500Ret.TS,end=c(2010,7))+ window(CurveSlope.TS,start=c(1986,2),end=c(2010,7)) + window(CreditSpread.TS,start=c(1986,2),end=c(2010,7)) + window(NAPM.TS,start=c(1986,2),end=c(2010,7)) + window(MCOILWTI.TS,start=c(1986,2),end=c(2010,7)) , family=binomial(link="probit")) predict(recession.probModel,window(full.TS,start=c(2010,8)), type="response") The GLM model works fine, but the predict function throws the following error: Error in window.default(FEDFUNDS.TS, start = c(1986, 2), end = c(2010, : 'start' cannot be after 'end' In addition: Warning message: In window.default(FEDFUNDS.TS, start = c(1986, 2), end = c(2010, : 'end' value not changed Can someone let me know why I am getting this error and what to do about it? Thanks, Bryson -- View this message in context: http://r.789695.n4.nabble.com/out-of-sample-predictions-using-a-probit-model-tp3772175p3772175.html Sent from the R help mailing list archive at Nabble.com. ______________________________________________ R-help@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.
