Fitting the curve problem

I try to fit an s-shape curve with the function that I define. But I only got a bad fit with the constant. Attached file is the data file and my scripts. Any advice will be appreciated.

[Image Analyst]

For our convenience can you attach a screenshot of the data plotted?

[Yumeng Yin]

Answer 1

[David Hill]

[fitcurve,gof]=fit(duration_fit,y_duration_fit,'fourier5');%fourier5 does not look too bad

[Yumeng Yin]

Thanks a lot, David. That is another new idea for me to think about. But I need to use my own defined equation to fit it so that I could extract the parameter that I define. I am wondering if my setting is not complete and it leads to a bad fitting to my equation.