< <
>
< <
= 0, other size
Note: Put
1
, 0, 0,
2
a b c = = = then the above distribution reduces to normal
distribution, where t = 3.14159., and e = 2.71828..
2) Sine Distribution:-
The Continuous random variable x which is distributed according to the probably
law:
f(x) =
2 2
1 1
.
a x
t
;
0
a x a
a
< <
>
= 0, otherwise
3) Sec Distribution:
The continuous random variable x which is distributed according to the probably law
( )
2 2
2 1
.
a
f x
x x a
t
=
; 0<a<x<
= 0, otherwise
4) Sine 2 distribution :
( )
2 2
2
2
. f x a x
a t
= ; -a< x <a , a>0
Proofs:
1 The Generalized Normal Distribution is a pdf because of the following
integral :
( )
2
2
( 4 )
4
.
a
b ac
ax bx c
a
e dx e
t
+ +
=
}
; a>0.
2
1 1 1
2 2
sin sin sin
a
a
a a
dx x a a
a a a
a x
( | | | | | |
= =
| | | (
\ . \ . \ .
}
= ( ) ( )
1 1
sin 1 sin 1
=
1 1
sin sin sin
2 2
t t
| | | | | |
| | |
\ . \ . \ .
=
2 2
t t
t + =
3.
1 1 1
2 2
0 0
1 1
.sec ( ) sec (0)
dx x
sec
a a a
x x a
( | |
( = =
| (
\ .
}
=
1
0
2 2 a a
t t (
=
(
5. Lambda Distribution:
A random variable x is said to have a Lambda distribution with parameters , , o if
its pdf is given by the probability law:
f(x; , , o ,a) =
2
2
1 log
. . ;
2
a x
a
o
o t
| |
|
\ .
a>0, >1, o >0,
- <x<,
- <
<
Proof:- It is known that
2 2
2 2
0
2
2
log
az az
dz dz
a
t
= =
} }
Note: Put = e, a =1, the above distribution reduces to normal distribution
f(x) =
2
1
2
1
.
2
x
e
o
o t
| |
|
\ .
6. Generalized Laplace Distribution:
A continuous random variable x is said to have generalized Laplace distribution with
parameters and if its pdf. is given by
f(x)=
| |
.
2
log
u x
a
a
< <
< <
> >
x
a 1 , 0
Proof: Taking u =
o
x
dx du = x=u( ) 1 +
} } }
+ = dx a dx a dx a
x x u x
. . .
0
) ( 0 | |
=
} }
+
0
0
. . du a du a
u u
=
0
0
log log
u u
a a
a a
( (
( (
= | |
a a a log
2
2 .
log
) 1 0 ( ) 0 1 (
log
= =
7. Purelog Distribution:
A random variable is said to have pure log distribution if its pdf is given by ;
f(x) =
) 1 (
) 1 log(
). ( log .
2 2
2 2
1
x b
x a
b
b a b
+
+ +
t
;
; 0
0
b a
b
x
<
>
< <
Proof:
}
+
+
0
2 2
2 2
) 1 (
) 1 log(
x b
x a
= ) log( .
b
b a
b
+ t
Goota funticon:
Define the integral as below:
) (n t = dx x a
n x
. .
1
0
}
a>1, n>0, ( log 1
e
=0)
) (n t =x
n-1
.
}
) 1 ( . n dx a
x
} }
dx dx a x
x n
). . (
2
=
}
dx a x
a
n
a
a
x
x n
x
n
. .
log
) 1 (
log
.
2 1
= ) 1 (
) log(
1
] . [
) log(
1
0
1
n
a
n
a x
a
x n
t
= ) 1 ( .
) log(
) 1 (
) 0 0 (
) log(
1
n
a
n
a
t
=
1
( 1)
log
e
n
n
a
t
=
0
1
. . ) , ( dx x a n a
n x
t
9. Beta distribution of 3
rd
kind:
A random variable is said to have Beta Distribution of 3
rd
kind if its pdf is given by ;
f(x) = ; ) 1 .( .
) , ( ). , (
) , ( ). log(
1 1
+
n m
x x
n a m a
n m a a
t t
t
1
1 0
0 ) , (
>
< <
>
a
x
n m
= 0, otherwise
10. Beta distribution of 4
th
kind:
A random variable is said to have Beta Distribution of 4
th
kind if its pdf is given by ;
f(x) = ;
) 1 (
.
) , ( ). , (
) , ( ). log(
1
n m
m
x
x
n a m a
n m a a
+
+
+
t t
t
1
0 ) , (
>
< <
>
a
x o
n m
where
}
=
0
1
. . ) , ( dx x a n a
n x
t
I . Prove that
(1/ 2)
log
e
a
t
t = ; a>1
Proof: We have
}
=
0
1
. . ) ( dx x a n
n x
t ; a>1, n>0
Put n=
1
2
,
1/ 2
0
1
( ) . .
2
t
a t dt t
=
}
Put t=u
2
, dt = 2u.du
2 2
1
2
2
0 0
1
( ) .( ) .2 . 2 .
2
u u
a u u du a du t
= =
} }
2 2
0 0
1
( ) 2 . 2 .
2
x y
a dx a dy t
= =
} }
2 2
( )
2
0 0
1
[ ( )] 4 .
2
x y
a dx dy t
+
=
} }
(Cartesian form)
Put x=r.cosu , y = r.sinu
x
2
+y
2
=r
2
, dxdy = r.dr.du
then
2
)] 2 / 1 ( [t = 4
} }
=
=
2
0
2
0
. . .
2
t t
u
u
r
r
d dr r a (polar form)
= 4 u
t t
u
d dr r a
r
r
. . .
2
0
2
0
2
(
(
(
} }
=
=
---------------- 1
Put r
2
= v => 2r.dr = dv
Then consider
}
0
. .
2
dr r a
r
=
=
}
0 0
log 2
1
.
2
1
a
a
dv a
v
v
=
0
0
1 1 1
2log 2log 2log
v
a a a
a a a
( ( = =
By 1, 4 )] 2 / 1 ( [
2
= t
} }
= =
= =
2
0
2
0
log log 2
4
.
log 2
1
t
u
t
u
t
u u
a
d
a
d
a
=> (1/ 2)
log
e
a
t
t = ; a>1