Investment Studio > Expressions > Functions > Statistical > FDIST
float fdist(float x, integer degrees_of_freedom_numerator, integer degrees_of_freedom_denominator, boolean cumulative = TRUE)
Returns Snedecor's F probability function.
x must be >= 0.
degrees_of_freedom_numerator and degrees_of_freedom_denominator must be >= 1.
If cumulative = TRUE, the one-tailed CDF (Cumulative Distribution Function) is returned (equal to the probability that a stochastic variable with F distribution is >= x); otherwise, the PDF (Probability Density Function) is returned. If cumulative is omitted, it defaults to TRUE.
Mathematically, the F PDF is defined as the PDF of the stochastic function
| u_numerator / degrees_of_freedom_numerator | |
| F(u_numerator, u_denominator) = | ¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾¾ |
| u_denominator / degrees_of_freedom_denominator |
where u_numerator, u_denominator are independent, c2-distributed stochastic variables (see chidist) with degrees_of_freedom_numerator and degrees_of_freedom_denominator degrees of freedom, respectively. It can be interpreted as the probability function of the ratio of the variances of two independent data sets, and is used to compute the probability of measuring a given ratio, typically for the purpose of hypothesis testing (see ftest).
Example
The probability of measuring a ratio >= 1.5 between the variances of a data set with 2 degrees of freedom and a a data set with 3 degrees of freedom is
=fdist(1.5, 2, 3)
» 35.3%.