.BG
.FN F
.FN df
.FN pf
.FN qf
.FN rf
.TL
F Distribution
.CS
df(q, df1, df2)
pf(q, df1, df2)
qf(p, df1, df2)
rf(n, df1, df2)
.AG q
vector of (positive) quantiles.
Missing values (`NA's) are allowed.
.AG p
vector of probabilities.
Missing values (`NA's) are allowed.
.AG n
sample size.
If `length(n)' is larger than 1, then `length(n)' random values are returned.
.AG df1
vector of degrees of freedom for numerator.
.AG df2
vector of degrees of freedom for denominator.
.RT
density (`df'),
probability (`pf'),
quantile (`qf'), or
random sample (`rf')
for the F-distribution with degrees of freedom `df1' and `df2'.
Values are computed to single precision accuracy only.
.SP
.Tr
The density function is given by
.sp
.EQ
delim $$
f sub { d sub 1 , d sub 2 } ( q ) ~=~
{ GAMMA ( { d sub 1 + d sub 2 } over 2 ) } over
{ GAMMA ( d sub 1 over 2 ) GAMMA ( d sub 2 over 2 ) }
{ q sup { d sub 1 - 1 } } over
{ ( 1 + q ) sup { \(12 ( d sub 1 + d sub 2 ) } }
~~~ q ~>~ 0 , ~~ d sub 1 , d sub 2 ~>~ 0,
.EN
.sp
where $d sub 1$ and $d sub 2$ are the `df1' and `df2' parameters.
.En
.SE
The function `rf' causes creation of the dataset `.Random.seed' if it does
not already exist, otherwise its value is updated.
.EX
rf(10,5,15) #sample of 10 with 5 and 15 degrees of freedom
.KW distribution
.EQ
delim off
.EN
.WR
