But when I try to use my own data, it doesn't work. The fit is nowhere near the data. What am I missing?
I have already tried: -cutting off chunks of (almost zero) data from the response data before and after the response peak -shifting the response data on the x-axis so that its peak aligns with the peak of the output signal -shifting signal data on the y-axis such that it aligns with the response data at y=0 -increasing the initial parameter value of A=7000
Any hints would be greatly appreciated!
Edit: I shifted the response (IRF) a little better, now the fit seems better, but the fit report says that the fit didn't converge. What can I do better?
Perhaps a two-term exponential function will work better with your data. Also try removing (or masking) the far right end of the data where it drops abruptly.
We are happy to try if you can send us your raw data and your response function data. Please send your file to tech@originlab.com and mention this post in your e-mail.