[R] Is there dpois equivalent for zero-inflated Poisson?
Thierry Onkelinx
thierry.onkelinx at inbo.be
Tue Mar 22 13:30:44 CET 2016
zero = proportion of zero inflation part
lamba = expected value of poisson part
There was a typo in the distribution. It should multiple by (1 - zero)
instead of divide by it.
dzeroinflpois <- function(x, lambda, zero){
ifelse(x == 0, zero, 0) + dpois(x, lambda) * (1 - zero)
}
ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature and
Forest
team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance
Kliniekstraat 25
1070 Anderlecht
Belgium
To call in the statistician after the experiment is done may be no more
than asking him to perform a post-mortem examination: he may be able to say
what the experiment died of. ~ Sir Ronald Aylmer Fisher
The plural of anecdote is not data. ~ Roger Brinner
The combination of some data and an aching desire for an answer does not
ensure that a reasonable answer can be extracted from a given body of data.
~ John Tukey
2016-03-22 13:25 GMT+01:00 Matti Viljamaa <mviljamaa op kapsi.fi>:
> Could you clarify what are the parameters and why it’s formulated that way?
>
> -Matti
>
> On 22 Mar 2016, at 14:17, Thierry Onkelinx <thierry.onkelinx op inbo.be>
> wrote:
>
> Dear Matti,
>
> What about this?
>
> dzeroinflpois <- function(x, lambda, zero){
> ifelse(x == 0, zero, 0) + dpois(x, lambda) / (1 - zero)
> }
> plot(x, dzeroinflpois(x, lambda = 10, zero = 0.2), type = "l")
>
>
>
> ir. Thierry Onkelinx
> Instituut voor natuur- en bosonderzoek / Research Institute for Nature and
> Forest
> team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance
> Kliniekstraat 25
> 1070 Anderlecht
> Belgium
>
> To call in the statistician after the experiment is done may be no more
> than asking him to perform a post-mortem examination: he may be able to say
> what the experiment died of. ~ Sir Ronald Aylmer Fisher
> The plural of anecdote is not data. ~ Roger Brinner
> The combination of some data and an aching desire for an answer does not
> ensure that a reasonable answer can be extracted from a given body of data.
> ~ John Tukey
>
> 2016-03-22 13:04 GMT+01:00 Matti Viljamaa <mviljamaa op kapsi.fi>:
>
>> I’m doing some optimisation that I first did with normal Poisson (only
>> parameter theta was estimated), but now I’m doing the same with a
>> zero-inflated Poisson model which
>> gives me two estimated parameters theta and p (p is also pi in some
>> notation).
>>
>> My question is, is there something equivalent to dpois that would use
>> both of the parameters (or is the p parameter possibly unnecessary)?
>>
>> I’m calculating the “fit” of the Poisson model
>>
>> i.e. like
>>
>> x = c(0,1,2,3,4,5,6)
>> y = c(3062,587,284,103,33,4,2)
>> fit1 <- sum(y)*dpois(x, est_theta)
>>
>> and then comparing fit1 to the real observations.
>> [[alternative HTML version deleted]]
>>
>> ______________________________________________
>> R-help op r-project.org mailing list -- To UNSUBSCRIBE and more, see
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide
>> http://www.R-project.org/posting-guide.html
>> <http://www.r-project.org/posting-guide.html>
>> and provide commented, minimal, self-contained, reproducible code.
>
>
>
>
[[alternative HTML version deleted]]
More information about the R-help
mailing list