We have some kinetic lifetime data that we are trying to model by the convolution of a gaussian instrument response function with a stretched exponential function for the signal. For reference the stretched exponential expression is given below.
y = Y0 + A * exp[-(x/t0)^b]
Unfortunately, the kinetic data we measure cannot be satisfactorily fit to either an exponential or biexponential function, and there is no physical basis for a triexponential model to be used.
I have tried fitting the signal and response together with the readily available convolution functions on the gaussian forums and cannot obtain a satisfying fit.
I have also followed the tutorial found here to perform the fit to a stretched exponential function in the worksheet with little success.