nctinv
statistics: x = ncx2inv (p, df, mu)
Inverse of the non-central -cumulative distribution function (iCDF).
For each element of p, compute the quantile (the inverse of the CDF) of the noncentral -distribution with df degrees of freedom and noncentrality parameter mu. The size of x is the common size of p, df, and mu. A scalar input functions as a constant matrix of the same size as the other inputs.
nctinv uses Newton’s method to converge to the solution.
Further information about the noncentral -distribution can be found at https://en.wikipedia.org/wiki/Noncentral_t-distribution
See also: nctcdf, nctpdf, nctrnd, nctstat, tinv
Source Code: nctinv
Plot various iCDFs from the noncentral T distribution
p = 0.001:0.001:0.999;
x1 = nctinv (p, 1, 0);
x2 = nctinv (p, 4, 0);
x3 = nctinv (p, 1, 2);
x4 = nctinv (p, 4, 2);
plot (p, x1, '-r', p, x2, '-g', p, x3, '-k', p, x4, '-m')
grid on
ylim ([-5, 5])
legend ({'df = 1, μ = 0', 'df = 4, μ = 0', ...
'df = 1, μ = 2', 'df = 4, μ = 2'}, 'location', 'northwest')
title ('Noncentral T iCDF')
xlabel ('probability')
ylabel ('values in x')
Compare the noncentral T iCDF with MU = 1 to the T iCDF with the same number of degrees of freedom (10).
p = 0.001:0.001:0.999;
x1 = nctinv (p, 10, 1);
x2 = tinv (p, 10);
plot (p, x1, '-', p, x2, '-');
grid on
ylim ([-5, 5])
legend ({'Noncentral T(10,1)', 'T(10)'}, 'location', 'northwest')
title ('Noncentral T vs T quantile functions')
xlabel ('probability')
ylabel ('values in x')