Re: [R] approximation function

Thanks for your reply. I know the exact values. The purpose is to find out the distribution function if it is exponential, linear, etc. Best, Carol --- On Mon, 4/6/09, Ravi Varadhan wrote: From: Ravi Varadhan Subject: Re: [R] approximation function To: wht_...@yahoo.com Cc: r-h

Re: [R] approximation function

Hi, At the sampling points, do you know the function value "exactly" or you only observe it with "noise"? If it is the former, you can use an interpolation scheme, such as, for example, interpSpline() in "splines" package. If it is the latter, you can use a smoother, such as, for example, smo

Re: [R] approximation function

Hi, This is a rather broad question. The function lowess() might be a good place to start, I guess (although I'm not sure whether this fits in your definition of "approximation functions"...). Best of luck, Luc carol white wrote: Hi, Having a set of values (non-time series data), what a