Can any of you guys or gals solve a curve-filtting problem for me?
Without going into numbers here, I have twelve data points. Also, I know the mean of data points 13 to 24, and I kow the mean of data points 25 to 36. The only constraints are that the curve is monotonically descreasing, is always greater than or equal to zero, and is hypothesised to be asymptotic to some (unknown) non-zero value.
I'd like a reasonable fit, doesn't have to be amazingly close.