how can I run the distributionFitter multiple times at once?

Dear All, I generated 40 data sets (data1 to data40), i need to find the normal distribution parameters for each data set, I use the distributionFitter command and it works well, my question is how can i please do that at one time and save the parameters values for each data set? i mean i want to save time instead of running the fitter 40 times? is there any function for this?

Answer 1

[Star Strider]

I would use the fitdist (link) function instead to fit your data.

[MAHMOUD ALZIOUD]

ok I just used the following code but it gives me Undefined function or variable 'tLocationScale'. I do not need the normal distribution anymore, i will use the t location scale distribution, what must the syntax be?

pd = fitdist(data1, tLocationScale)

[MAHMOUD ALZIOUD]

i just fixed it and it worked thank you sir

[Star Strider]

As always, my pleasure.

[Pr-Gh]

I am facing a similar problem of having to fit a given probability distribution to large number of different unique data sets so how can I do it inone go.

The answer did not become clear . I do not find the function fitdist(link) in matlab.

[Star Strider]

@Pr-Gh — Look again for fitdist since it still very much exists!

Note that ‘(link)’ is not part of the function name!

[Pr-Gh]

Fitdist is there but does it require a loop( such as" for" ) to read the large number of data sets or otherwise...that part is not clear

[Star Strider]

@Pr-Gh — It is very clear!

From, the documentation:

‘pd = fitdist(x,distname) creates a probability distribution object by fitting the distribution specified by distname to the data in column vector x.’ (Emphasis added)

The argument has to be a vector.