Definition Student's t-distribution
1 definition
1.1 probability density function
1.2 cumulative distribution function
1.3 special cases
definition
probability density function
student s t-distribution has probability density function given by
f
(
t
)
=
Γ
(
ν
+
1
2
)
ν
π
Γ
(
ν
2
)
(
1
+
t
2
ν
)
−
ν
+
1
2
,
{\displaystyle f(t)={\frac {\gamma ({\frac {\nu +1}{2}})}{{\sqrt {\nu \pi }}\,\gamma ({\frac {\nu }{2}})}}\left(1+{\frac {t^{2}}{\nu }}\right)^{\!-{\frac {\nu +1}{2}}},\!}
where
ν
{\displaystyle \nu }
number of degrees of freedom ,
Γ
{\displaystyle \gamma }
gamma function. may written as
f
(
t
)
=
1
2
ν
b
(
1
2
,
ν
2
)
(
1
+
t
2
ν
)
−
ν
+
1
2
,
{\displaystyle f(t)={\frac {1}{2{\sqrt {\nu }}\,\mathrm {b} ({\frac {1}{2}},{\frac {\nu }{2}})}}\left(1+{\frac {t^{2}}{\nu }}\right)^{\!-{\frac {\nu +1}{2}}}\!,}
where b beta function. in particular integer valued degrees of freedom
ν
{\displaystyle \nu }
have:
for
ν
>
1
{\displaystyle \nu >1}
even,
Γ
(
ν
+
1
2
)
ν
π
Γ
(
ν
2
)
=
(
ν
−
1
)
(
ν
−
3
)
⋯
5
⋅
3
2
ν
(
ν
−
2
)
(
ν
−
4
)
⋯
4
⋅
2
⋅
{\displaystyle {\frac {\gamma ({\frac {\nu +1}{2}})}{{\sqrt {\nu \pi }}\,\gamma ({\frac {\nu }{2}})}}={\frac {(\nu -1)(\nu -3)\cdots 5\cdot 3}{2{\sqrt {\nu }}(\nu -2)(\nu -4)\cdots 4\cdot 2\,}}\cdot }
for
ν
>
1
{\displaystyle \nu >1}
odd,
Γ
(
ν
+
1
2
)
ν
π
Γ
(
ν
2
)
=
(
ν
−
1
)
(
ν
−
3
)
⋯
4
⋅
2
π
ν
(
ν
−
2
)
(
ν
−
4
)
⋯
5
⋅
3
⋅
{\displaystyle {\frac {\gamma ({\frac {\nu +1}{2}})}{{\sqrt {\nu \pi }}\,\gamma ({\frac {\nu }{2}})}}={\frac {(\nu -1)(\nu -3)\cdots 4\cdot 2}{\pi {\sqrt {\nu }}(\nu -2)(\nu -4)\cdots 5\cdot 3\,}}\cdot \!}
the probability density function symmetric, , overall shape resembles bell shape of distributed variable mean 0 , variance 1, except bit lower , wider. number of degrees of freedom grows, t-distribution approaches normal distribution mean 0 , variance 1. reason
ν
{\displaystyle {\nu }}
known normality parameter .
the following images show density of t-distribution increasing values of
ν
{\displaystyle \nu }
. normal distribution shown blue line comparison. note t-distribution (red line) becomes closer normal distribution
ν
{\displaystyle \nu }
increases.
cumulative distribution function
the cumulative distribution function can written in terms of i, regularized incomplete beta function. t > 0,
f
(
t
)
=
∫
−
∞
t
f
(
u
)
d
u
=
1
−
1
2
i
x
(
t
)
(
ν
2
,
1
2
)
,
{\displaystyle f(t)=\int _{-\infty }^{t}f(u)\,du=1-{\tfrac {1}{2}}i_{x(t)}\left({\tfrac {\nu }{2}},{\tfrac {1}{2}}\right),}
where
x
(
t
)
=
ν
t
2
+
ν
.
{\displaystyle x(t)={\frac {\nu }{t^{2}+\nu }}.}
other values obtained symmetry. alternative formula, valid
t
2
<
ν
{\displaystyle t^{2}<\nu }
, is
∫
−
∞
t
f
(
u
)
d
u
=
1
2
+
t
Γ
(
1
2
(
ν
+
1
)
)
π
ν
Γ
(
ν
2
)
2
f
1
(
1
2
,
1
2
(
ν
+
1
)
;
3
2
;
−
t
2
ν
)
,
{\displaystyle \int _{-\infty }^{t}f(u)\,du={\tfrac {1}{2}}+t{\frac {\gamma \left({\tfrac {1}{2}}(\nu +1)\right)}{{\sqrt {\pi \nu }}\,\gamma \left({\tfrac {\nu }{2}}\right)}}\,{}_{2}f_{1}\left({\tfrac {1}{2}},{\tfrac {1}{2}}(\nu +1);{\tfrac {3}{2}};-{\tfrac {t^{2}}{\nu }}\right),}
where 2f1 particular case of hypergeometric function.
for information on inverse cumulative distribution function, see quantile function#student s t-distribution.
special cases
certain values of
ν
{\displaystyle \nu }
give simple form.
ν
=
1
{\displaystyle \nu =1}
distribution function:
f
(
t
)
=
1
2
+
1
π
arctan
(
t
)
.
{\displaystyle f(t)={\tfrac {1}{2}}+{\tfrac {1}{\pi }}\arctan(t).}
density function:
f
(
t
)
=
1
π
(
1
+
t
2
)
.
{\displaystyle f(t)={\frac {1}{\pi (1+t^{2})}}.}
see cauchy distribution
ν
=
2
{\displaystyle \nu =2}
distribution function:
f
(
t
)
=
1
2
+
t
2
2
+
t
2
.
{\displaystyle f(t)={\tfrac {1}{2}}+{\frac {t}{2{\sqrt {2+t^{2}}}}}.}
density function:
f
(
t
)
=
1
(
2
+
t
2
)
3
2
.
{\displaystyle f(t)={\frac {1}{\left(2+t^{2}\right)^{\frac {3}{2}}}}.}
ν
=
3
{\displaystyle \nu =3}
density function:
f
(
t
)
=
6
3
π
(
3
+
t
2
)
2
.
{\displaystyle f(t)={\frac {6{\sqrt {3}}}{\pi \left(3+t^{2}\right)^{2}}}.}
ν
=
∞
{\displaystyle \nu =\infty }
density function:
f
(
t
)
=
1
2
π
e
−
t
2
2
.
{\displaystyle f(t)={\frac {1}{\sqrt {2\pi }}}e^{-{\frac {t^{2}}{2}}}.}
see normal distribution
^ john kruschke (2014), doing bayesian data analysis, academic press; 2 edition. isbn 0124058884
^ cite error: named reference jkb invoked never defined (see page).
Comments
Post a Comment