Dear members, I am using the forecast package for arfima modelling. THe following is the code:
> arfima(ygrpch(OHLCDataEP[[1]])) Call: arfima(y = ygrpch(OHLCDataEP[[1]])) *** Warning during (fdcov) fit: unable to compute correlation matrix; maybe change 'h' Coefficients: d ma.ma1 0.1745253 0.1600765 sigma[eps] = 34.4493 a list with components: [1] "log.likelihood" "n" "msg" "d" "ar" [6] "ma" "covariance.dpq" "fnormMin" "sigma" "stderror.dpq" [11] "correlation.dpq" "h" "d.tol" "M" "hessian.dpq" [16] "length.w" "residuals" "fitted" "call" "x" [21] "series" What does the warning say? Is it Ok to proceed with it if the correlation matrix can't be caclulated? I just want the prediction, and forecast() is working fine: > forecast(arfima(ygrpch(OHLCDataEP[[1]])), h = 1) Point Forecast Lo 80 Hi 80 Lo 95 Hi 95 188 26.01569 -17.96402 69.99539 -41.24547 93.27684 I just want the point forecast. Is there any problem with arfima() as I am using it? THanking you, Yours sincerely, AKSHAY M KULKARNI [[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see 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.