Info
この質問は閉じられています。 編集または回答するには再度開いてください。
Fitting a distribution to an array containing zeros
2 ビュー (過去 30 日間)
古いコメントを表示
I have data from stimulation of subjects with different intensities (say we have 54 different intensities), and what follows is the percentage of them recognizing the respective stimulation:
x = [0 0 0 0.50 0 0 0 0 0 0 0.5 0 0 0 0 0 0 0 0.125000000000000 0 0.333333333333333 0 0 0.111111111111111 0 0.428571428571429 0 0.285714285714286 0.166666666666667 0 0.1 0 0.400000000000000 0.5 0.4 0.25 0.6 0.727272727272727 0.714285714285714 0.25 0.666666666666667 0.777777777777778 1 0.75 0 1 0.9375 1 1 1 1 1 0.92 0.92]
Say the first index is the weakest stimulation, and the last index the strongest, as visible the stronger the stimulation the more likely the subject recognizes it. I want to fit now a distribution to these values, to get something called a psychophysical curve, usually a logistic function (looks like [this][1]).
What I have tried is:
pd = fitdist(x,distribution);
but this throws an error, I assume because of the 0's in the x array. What could I do alternatively?
[1]: http://www.cns.nyu.edu/~david/courses/perception/lecturenotes/psychophysics/psychophysics%20slides/Slide7.jpg
1 件のコメント
Jos (10584)
2017 年 12 月 13 日
What is the exact error you get?
What is the exact distribution you are fitting?
回答 (0 件)
この質問は閉じられています。
参考
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!