From Wikipedia, the free encyclopedia
An exponentially modified Gaussian distribution is often encountered in modelling chromatographic peaks.[ 1] The general form of the( probability distribution) equation is given below.[ 2]
f
(
x
)
=
y
0
+
A
t
0
e
x
p
(
1
2
(
w
t
0
)
2
−
x
−
x
c
t
0
)
∫
−
∞
z
1
2
e
x
p
(
−
y
2
z
)
d
y
{\displaystyle f(x)=y_{0}+{\frac {A}{t_{0}}}exp\left({\frac {1}{2}}\left({\frac {w}{t_{0}}}\right)^{2}-{\frac {x-x_{c}}{t_{0}}}\right)\int _{-\infty }^{z}{\frac {1}{\sqrt {2}}}exp\left({\frac {-y^{2}}{z}}\right)dy}
or
f
(
x
)
=
y
0
+
A
t
0
e
x
p
(
1
2
(
w
t
0
)
2
−
x
−
x
c
t
0
)
(
1
2
+
1
2
e
r
f
(
z
2
)
)
{\displaystyle f(x)=y_{0}+{\frac {A}{t_{0}}}exp\left({\frac {1}{2}}\left({\frac {w}{t_{0}}}\right)^{2}-{\frac {x-x_{c}}{t_{0}}}\right)\left({\frac {1}{2}}+{\frac {1}{2}}erf\left({\frac {z}{\sqrt {2}}}\right)\right)}
y
0
{\displaystyle y_{0}}
= the initial value
A
{\displaystyle A}
= the amplitude
x
c
{\displaystyle x_{c}}
= the center of the peak
w
{\displaystyle w}
= the width of the peak
t
0
{\displaystyle t_{0}}
= the modification factor (skewness,
t
0
>
0
{\displaystyle t_{0}>0}
)
z
=
x
−
x
c
w
−
w
t
0
{\displaystyle z={\frac {x-x_{c}}{w}}-{\frac {w}{t_{0}}}}
e
r
f
(
z
2
)
{\displaystyle erf\left({\frac {z}{\sqrt {2}}}\right)}
= the error function
The area under the curve (cumulative distribution) is given by[ 3]
F
(
x
)
=
A
2
t
0
exp
(
w
2
2
t
0
2
+
x
c
−
x
t
0
)
[
e
r
f
(
x
−
x
c
2
w
−
w
2
t
0
)
+
1
]
+
c
0
{\displaystyle F(x)={\frac {A}{2t_{0}}}\exp \left({\frac {w^{2}}{2t_{0}^{2}}}+{\frac {x_{c}-x}{t_{0}}}\right)\left[\mathrm {erf} \left({\frac {x-x_{c}}{{\sqrt {2}}w}}-{\frac {w}{{\sqrt {2}}t_{0}}}\right)+1\right]+c_{0}}
which is the same as...
F
(
x
)
=
c
0
+
A
t
0
e
x
p
(
1
2
(
w
t
0
)
2
−
x
−
x
c
t
0
)
(
1
2
+
1
2
e
r
f
(
z
2
)
)
{\displaystyle F(x)=c_{0}+{\frac {A}{t_{0}}}exp\left({\frac {1}{2}}\left({\frac {w}{t_{0}}}\right)^{2}-{\frac {x-x_{c}}{t_{0}}}\right)\left({\frac {1}{2}}+{\frac {1}{2}}erf\left({\frac {z}{\sqrt {2}}}\right)\right)}
References