On 20-02-2013, at 21:13, Rolf Turner <rolf.tur...@xtra.co.nz> wrote:
> > > This question doesn't make a lot of sense to me. What do > you expect/want your "XXX" values to be? Well the OP wants some sort of interpolation. > If you only know > that the value for the first quarter was 100, you cannot infer > the individual values for January, February and March. All > you know is that these values sum to 100. And if you a value for the second quarter you can interpolate. The quarterly data could be stock data such as a money stock. In that case the new values wouldn't sum to e.g. 100. The quarterly data would be located at the start or end of a quarter. The OP didn't specify if the data were stocks or flows. The spline method the OP used is a perfectly reasonable route to take. If you have a model for the low frequency data there is a very new package for temporal disagggregation: tempdisagg. CRAN Task View: Time Series Analysis provide more timeseries stuff. I think that package zoo could also help but I haven't been able to construct an example. For the OP: are your quarterly data stocks of flows? If they are flows you probably want the interpolated monthly values to sum to the quarterly value. If they are stocks you should specify if your quarterly values are end-of-quarter or start-of-quarter. Berend ______________________________________________ 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.
