Use DistributionFitTest. Here is an example I ran with some random data (certainly does not fit the Burr distribution).

data = RandomReal[{0, 1}, 400];
test = DistributionFitTest[data, SinghMaddalaDistribution[2, 2.5, 3],"HypothesisTestData"];
test["TestDataTable", All]

The Burr distribution is called the Singh-Maddala distribution in Mathematica (and in many other places). The output of the last command will be a table with the results (statistics and p-values) of applying the tests Mathematica considers appropriate for the data used (Anderson-Darling, Cramer-von Mises, Kolmogorov-Smirnov, etc.).

Notice that one needs to enter the distribution parameters one wants to test. That is generally necessary, except for distributions that are 'nicely' scalable, such as the normal distribution. If you think some distribution of this type should fit your data, but you are not sure what parameters would work, then you are dealing with a different problem and you should use a different tool (in that case I would use FindDistributionParameters).

Best,

OL.

I am interested in finding the goodness of fit test. For parameter testing it gives no answer.Plz check where I commit the mistake.

Thank you Sean...I have applied your suggested approach. But It gives no result,Tell me where is error as I have attach the file .

Attachments:

Thank you Sean...I have applied your suggested approach. But It gives no result,Tell me where is error as I have attach the file .

Attachments:

You surely are a bit scarce with providing information on what you are trying to solve, what kind of data, and what functions you have used/tried. So I will give a general answer:

If you used LinearModelFit or NonlinearModelFit you can query many options, the simplest being the residuals, but there are also ANOVA tables and many more values you can obtain from your fits...

Have a look at the help of those.