I was playing around with the nnetar function in the forecast package, trying to generate a 3-year forecast (36 months), the R fit result is shown below: #Neural Nets Fitting and Forecast> > fit<-nnetar(condataset$ConPcums97) > fitSeries: condataset$ConPcums97 Model: NNAR(13) Call: nnetar(x = condataset$ConPcums97) Average of 20 networks, each of which is a 13-7-1 network with 106 weights options were - linear output units sigma^2 estimated as 5.34e+12 Now, R is sending me the following error message: x<-forecast(fit, h=36) Error in predict.nnet(X[[1L]], ...) : missing values in 'x'> > forecast(fit)Error in predict.nnet(X[[1L]], ...) : missing values in 'x'> x<-forecast(nnetar(condataset$ConPcums97), h=36)Error in predict.nnet(X[[1L]], ...) : missing values in 'x' I am attaching the same data I used to try the nnetar function. Can you give me some guidance or tell me what could be done to generate the forecast with this function? Best regards, Paul