Hi All, By a production curve I mean for example the output of a mine, peak oil production or the yield of a farm over time within the same season. It is this last example that we should take as the prototypical case.
What I would like to do is to fit a curve that inherits qualities of the discrete production data (such as area of the curve equaling the total production for the season). Fitting a curve with least squares (such as a Gaussean or Hubbert) presents some issues (with regards to accuracy of inherited features). My next logical attempt would be to fit a sum of curves, such as a Fourier or Wavelet sum. Perhaps there is something simpler or more flexible in the way I am thinking? My question is: 1. What would be an effective approach be to fit generalised production curves? 2. If a Wavelet sum is one of the best approaches, what would be a good way of implementing such curve fitting (including calculated coefficients) in R? 3. Is there anything else or another way that I should rather be thinking about this instead? Best regards Phillip-Jan van Zyl MSc Mathematics, Stellenbosch ______________________________________________ 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.
