There is a discrete sample of 501 elements (0.4042, 0.5967, 0.2124, 0.4576, etc. ...). To analyze the compliance of this sample with the normal distribution law (Gaussian distribution).

I looked through many articles, but everywhere, as it is, it is described differently and even in theory. ver. I am rather weak ... Who will help with the solution algorithm? Or maybe someone had a similar job?

  • one
    This is done very simply with the help of statistical analysis of the sample, asymmetry and kurtosis are found, which are then compared with the dispersion values ​​of these same asymmetries and kurtosis with some coefficients; I just don’t understand, and here C ++ and especially Qt - MorkOFF
  • I just write the program on Qt, so I added tags ... if someone explains in more detail how to solve the problem I will be very grateful ... - LANSELOT
  • one
    We used Pearson's criterion. The gap is divided into intervals, the theoretical and experimental probability of falling into the interval is considered, summed by the formula and compared by a table value - Oleg Nechitailo
  • Empirical frequencies are the number of hits in each of the intervals, and theoretical frequencies are the frequencies that “should be” if the random variable were normally distributed. I understand correctly? How to calculate theoretical frequencies in C ++? There seems to be no way without a function (integral) of the normal distribution. - LANSELOT
  • That's right. You can use the Laplace function for counting. Take either the interpolation on the table, or decompose in a row. - Oleg Nechitaylo

0