Dear community,
I am currently trying to model the following problem in STATA:
I want to model asset returns (which I assume to be normally distributed). These asset returns can be affected by shocks (where each shock is normally distributed). Now my problem is that I want to model that the number of these shocks within a certain period (for example 3 months) is poisson-distributed. The idea was to model this by programming a MLE-maximization, but the problem is that I don't know how to model the poisson-distributed variable, as it is still part of my log-likelihood function.
this is the status quo (where x is the poisson-distributed variable):
`lnf' = ln*[exp([(-`lambda')*(`lambda'^x)/x!]*2π*[(`sigma2'+x*`gamma2')^-0.5]*[exp(-0.5(($ML_y1-(`alpha'+x*`theta')^2)/(`sigma2'+x*`gamma2')]]
I would be very happy if someone helped me out.
Airmir
I am currently trying to model the following problem in STATA:
I want to model asset returns (which I assume to be normally distributed). These asset returns can be affected by shocks (where each shock is normally distributed). Now my problem is that I want to model that the number of these shocks within a certain period (for example 3 months) is poisson-distributed. The idea was to model this by programming a MLE-maximization, but the problem is that I don't know how to model the poisson-distributed variable, as it is still part of my log-likelihood function.
this is the status quo (where x is the poisson-distributed variable):
`lnf' = ln*[exp([(-`lambda')*(`lambda'^x)/x!]*2π*[(`sigma2'+x*`gamma2')^-0.5]*[exp(-0.5(($ML_y1-(`alpha'+x*`theta')^2)/(`sigma2'+x*`gamma2')]]
I would be very happy if someone helped me out.
Airmir
