[SciPy-user] Combination of negative-exponential + normal distribution, how to tackle ?

[Stef Mientki]

thanks David,

David Huard wrote:
> Hi Stef,
>
> I'm not sure if that's really what you want to do, but you could find 
> the parameters maximizing the likelihood of a mixture of both 
> distributions:
>
> A * normal_likelihood(mu, sigma) + B * 
> negative_exponential_likelihood(params)
> A+B = 1
>
I think that's waht I'm looking for.
> Anyway, google "mixture model" for more info. There is a wikipedia 
> page about this where you might find more appropriate solutions to 
> your problem.
>
Oh, enought to spend the whole weekend ;-)
Thanks,
cheers,
Stef Mientki


Kamer van Koophandel - handelsregister 41055629  / Netherlands Chamber of Commerce - trade register 41055629