The triangular distribution (or Simpson distribution , after Thomas Simpson ) is a continuous probability distribution that is used in probability theory and statistics .
definition
The triangular distribution is defined by the probability density function defined on the interval
[
a
,
b
]
{\ displaystyle \ left [a, b \ right]}
f
(
x
)
=
{
2
(
x
-
a
)
(
b
-
a
)
(
c
-
a
)
,
if
a
≤
x
<
c
2
b
-
a
,
if
x
=
c
2
(
b
-
x
)
(
b
-
a
)
(
b
-
c
)
,
if
c
<
x
≤
b
.
{\ displaystyle f (x) = {\ begin {cases} {\ frac {2 (xa)} {(ba) (ca)}}, & {\ text {if}} a \ leq x <c \\ { \ frac {2} {ba}}, & {\ text {if}} x = c \\ {\ frac {2 (bx)} {(ba) (bc)}}, & {\ text {if}} c <x \ leq b. \ end {cases}}}
The parameters (minimum value), (maximum value) and (most likely value) determine the shape of the triangular distribution ( and ). The graph of the density function looks like a triangle and gives this distribution its name. The axis shows the density of the respective probability for a value .
a
{\ displaystyle a}
b
{\ displaystyle b}
c
{\ displaystyle c}
a
<
b
{\ displaystyle a <b}
a
≤
c
≤
b
{\ displaystyle a \ leq c \ leq b}
y
{\ displaystyle y}
x
∈
[
a
,
b
]
{\ displaystyle x \ in \ left [a, b \ right]}
properties
Distribution function
The distribution function
The distribution function is
F.
(
x
)
=
{
(
x
-
a
)
2
(
b
-
a
)
(
c
-
a
)
,
if
a
≤
x
<
c
c
-
a
b
-
a
,
if
x
=
c
1
-
(
b
-
x
)
2
(
b
-
a
)
(
b
-
c
)
,
if
c
<
x
≤
b
.
{\ displaystyle F (x) = {\ begin {cases} {\ frac {(xa) ^ {2}} {(ba) (ca)}}, & {\ text {if}} a \ leq x <c \\ {\ frac {ca} {ba}}, & {\ text {if}} x = c \\ 1 - {\ frac {(bx) ^ {2}} {(ba) (bc)}}, & {\ text {if}} c <x \ leq b. \ end {cases}}}
The inverse of the distribution function is
F.
-
1
(
y
)
=
{
a
+
y
(
b
-
a
)
(
c
-
a
)
,
if
0
≤
y
≤
(
c
-
a
)
(
b
-
a
)
b
-
(
b
-
a
)
(
b
-
c
)
(
1
-
y
)
,
if
(
c
-
a
)
(
b
-
a
)
≤
y
≤
1
{\ displaystyle F ^ {- 1} (y) = {\ begin {cases} a + {\ sqrt {y (ba) (ca)}}, & {\ text {if}} 0 \ leq y \ leq {\ frac {(ca)} {(ba)}} \\ b - {\ sqrt {(ba) (bc)}} {\ sqrt {(1-y)}}, & {\ text {if}} {\ frac {(ca)} {(ba)}} \ leq y \ leq 1 \ end {cases}}}
Expected value and median
The expected value of a triangular random variable is
X
{\ displaystyle X}
E.
(
X
)
=
a
+
b
+
c
3
.
{\ displaystyle \ operatorname {E} (X) = {\ frac {a + b + c} {3}}.}
For is the median given by
b
-
c
>
c
-
a
{\ displaystyle bc> ca}
m
{\ displaystyle m}
m
=
b
-
(
b
-
a
)
(
b
-
c
)
/
2
{\ displaystyle m = b - {\ sqrt {(ba) (bc) / 2}}}
. In this case the median is smaller than the expected value; d. H. the distribution is skewed to the right in the Pearson sense .
Variance
The variance of a triangular random variable results in
X
{\ displaystyle X}
Var
(
X
)
=
a
2
+
b
2
+
c
2
-
a
b
-
a
c
-
b
c
18th
=
(
a
-
b
)
2
+
(
b
-
c
)
2
+
(
a
-
c
)
2
36
.
{\ displaystyle \ operatorname {Var} (X) = {\ frac {a ^ {2} + b ^ {2} + c ^ {2} -ab-ac-bc} {18}} = {\ frac {( ab) ^ {2} + (bc) ^ {2} + (ac) ^ {2}} {36}}.}
Relationship to other distributions
Sum of uniformly distributed random variables
The sum of two identical, independent and continuously uniformly distributed random variables is triangularly distributed with , standard deviation , mean absolute deviation and interquartile range .
b
-
c
=
c
-
a
{\ displaystyle bc = ca}
6th
(
b
-
a
)
/
12
≈
0.204
(
b
-
a
)
{\ displaystyle {\ sqrt {6}} (ba) / 12 \ approx 0 {,} 204 (ba)}
(
b
-
a
)
/
6th
≈
0.167
(
b
-
a
)
{\ displaystyle (ba) / 6 \ approx 0 {,} 167 (ba)}
(
1
-
2
/
2
)
(
b
-
a
)
≈
0.293
(
b
-
a
)
{\ displaystyle (1 - {\ sqrt {2}} / 2) (ba) \ approx 0 {,} 293 (ba)}
Amount of the difference between uniformly distributed random variables
The amount of the difference between two identical, independent and continuously equally distributed random variables is triangularly distributed with .
|
X
1
-
X
2
|
{\ displaystyle | X_ {1} -X_ {2} |}
a
=
c
=
0
{\ displaystyle a = c = 0}
Trapezoidal distribution
The triangular distribution is a special case of the trapezoidal distribution .
Discrete triangular distribution
The continuous triangular distribution can be understood as the limit value of a discrete triangular distribution.
literature
Web links
Discrete univariate distributions
Continuous univariate distributions
Multivariate distributions
<img src="https://de.wikipedia.org/wiki/Special:CentralAutoLogin/start?type=1x1" alt="" title="" width="1" height="1" style="border: none; position: absolute;">