Test randomness from STD deviation

Hello

If I have a random seq of 50,000 data and all the data ranging between 0 to 255..then is it possible to test the true randomness by measuring std deviation......

    X = randi(256,1,50000)-1;

    P = histc(X,0:255);

what shall be next,how to measure if the data are uniformly random in nature by measuring std deviation....

Answer 1

[José-Luis]

No, the standard deviation by itself will tell you next to nothing about the nature of the distribution. Something like the two-sample Kolmogorov-Smirnov test might help you.

 doc kstest2

The problem remains of defining what true randomness is. If your numbers are generated by a computer, then what you get are pseudo-random numbers. For such numbers, the "randomness" breaks down once you have very many samples, or when you consider many dimensions. No all random numbers are created equal.