Creates an outcome with Poisson distribution with the chosen parameter.
Usage
Poisson(lambda, data, offset = as.matrix(data)^0)Arguments
- lambda
character: The name of the linear predictor associated with the rate (mean) parameter of the Poisson distribution. The parameter is treated as unknown and equal to the exponential of the associated linear predictor.
- data
numeric: The values of the observed data.
- offset
numeric: The offset at each observation. Must have the same shape as data.
Details
For evaluating the posterior parameters, we use the method proposed in ArtigokParametrico;textualkDGLM.
For the details about the implementation see ArtigoPacote;textualkDGLM.
See also
Other auxiliary functions for a creating outcomes:
Gamma(),
Multinom(),
Normal(),
summary.dlm_distr()
Examples
data <- c(AirPassengers)
level <- polynomial_block(rate = 1, D = 0.95, order = 2)
season <- harmonic_block(rate = 1, period = 12, D = 0.975)
outcome <- Poisson(lambda = "rate", data = data)
fitted.data <- fit_model(level, season,
AirPassengers = outcome
)
summary(fitted.data)
#> Fitted DGLM with 1 outcomes.
#>
#> distributions:
#> AirPassengers: Poisson
#>
#> ---
#> No static coeficients.
#> ---
#> See the coef.fitted_dlm for the coeficients with temporal dynamic.
#>
#> One-step-ahead prediction
#> Log-likelihood : -625.2975
#> Interval Score : 138.36923
#> Mean Abs. Scaled Error: 0.68377
#> ---
plot(fitted.data, plot.pkg = "base")
