Professional Documents
Culture Documents
Poisson Distribution
A Poisson distribution is given by:
Pr[Y y ]
e y
y!
, y 0,1,2..
Poisson Model
ln i x
'
i
E[ yi | xi ] i e
xi
Specification.
Thus:
E[ yi | xi ]
i i
xi
E[ yi | xi ] i e
xi'
Var [ yi | xi ]
Estimation
We estimate the model using MLE. The
Likelihood function is non linear:
n
ln L [i y x ln y i !] [e
i 1
'
i i
i 1
x i'
yi xi' ln yi !]
i 1
Estimation.
The Hessian of this function is:
n
L2
x i'
'
H
[ x i x i e ]
'
i 1
From this, we can get the asymptotic variancecovariance matrix of the ML estimator:
'
var asy ( ) [ [ x i' x i e xi ]] 1
i 1
i e
x i'
ik
log X ik
Diagnostic Testing
As we had mentioned before, a major assumption
of the Poisson model is:
E[ yi | xi ] i e 'x Var [ yi | xi ]
i
^
2
Diagnostic Testing
An alternative approach is by Wooldridge(1996)
which involves regressing the square of
standardized residuals-1 on the forecasted value
and testing alpha = 0 in the following test
(y )
equation
1
Diagnostic Testing
With miss-specification, the std errors will not be
consistent. We can compute robust std errors using
Huber/White (QML) option or GLM , which corrects the
std errors for miss-specification.
For Poisson, MLE are also QMLE
The respective std errors are:
1
varQML ( ) H
g g
i
'
i
And,
varGLM ( ) varML ( )
Where,
( yi i ) 2
1
N K i 1
i
2
Empirical Examples
Done in Eviews