memexp-6.0.exe simple dir3/data.003 2 dir3/irf.003 0 99. 0.2 0.9 1 1. 1.0 1 40 0 5.0D-4 0
The third test case analyzes data that were simulated as Poisson deviates corresponding to triexponential kinetics convolved with a known (i.e., measured) instrument response function (IRF). The temporal resolution is 1.0, and the zero-time shift is 0.5. No scattered light was added here. The simulated exponentials are of equal amplitude with lifetimes of 10, 30, and 100, respectively.
For a Poisson process, no preliminary estimate of uncertainties is needed; i.e., this analysis requires only one maximum entropy calculation (i.e., simple or invert mode) with IGNORE set to 2.
Both the noisy data and the noisy IRF must be input to MemExp. Before fitting the data, MemExp subtracts a baseline from the IRF and summarizes results in data.003.irf.ps.
Because the value of TSHIFT specified on the command line above (99.) is greater than 90, preliminary MEM calculations are performed at fixed values of the zero-time shift, and Brent's method of optimization is used to estimate the shift. All calculations are summarized in data.003.out, and these preliminary results are output in lines that begin with `Estimating TSHIFT>' and specify the Poisson deviance obtained for a given zero-time shift.
Once the zero-time shift has been estimated, the distributed fits are plotted in data.003.out.ps. In the fourth column, the Poisson deviance Pd is reported (IGNORE=2). In the first row of the fourth column, the temporal resolution of the data and the full width at half maximum (FWHM) of the IRF are marked by gray vertical lines and labeled as 'dt' and 'IRFW,' respectively.
The discrete fits are plotted in data.003.exp.ps. Because the value of VARYTS specified on the command line above is 1, the value of the zero-time shift is varied in these fits, along with the exponential amplitudes and lifetimes.