Evaluate the "goodness of fit" of a distribution fitting, when using the function "fitdist"

39 次查看（过去 30 天）

Sim 2023-6-14

评论： Mathieu NOE 2023-6-22

When I use the function fitdist to find the best distribution fitting, I would like to measure or assess the goodness of fitting.

What are the possible ways to measure the goodness of fitting of a distribution fitting, when employing the function fitdist?

% Example where I would like to assess the goodness of fit for this

% histogram:

rng default;

r = normrnd(10,10,1000,1);

histogram(r,'binwidth',1,'Normalization','pdf') % <-- my data

pd = fitdist(r,'normal'); % <-- fit the distribution to the data (fitdist)

xgrid = linspace(-30,100,100)';

pdfEst = pdf(pd,xgrid);

line(xgrid,pdfEst,'Linewidth',2,'color','r')

Sim 2023-6-22

Thanks a lot @Mathieu NOE, that's great! .....I did not think about the R2 for this case..... :-)

Mathieu NOE 2023-6-22

my pleasure !

在 Help Center 和 File Exchange 中查找有关 Half-Normal Distribution 的更多信息

Find the treasures in MATLAB Central and discover how the community can help you!

Translated by