Learn R Programming

tweedie (version 3.2.2)

ptweedie_series: Series Evaluation for the Tweedie Distribution Function

Description

Evaluates the distribution function (df) for Tweedie distributions with \(1 < p < 2\) using an infinite series, for given values of the dependent variable y, the mean mu, dispersion phi, and power parameter power. Not usually called by general users, but can be in the case of evaluation problems.

Usage

ptweedie_series(q, power, mu, phi, lower.tail = TRUE, log.p = FALSE, 
                       verbose = FALSE, details = FALSE, max.terms = 1e6)

ptweedie.series(q, power, mu, phi, verbose = FALSE, details = FALSE)

Value

A numeric vector of probabilities (or log-probabilities, if log.p = TRUE).

Arguments

q

vector of quantiles.

power

the power parameter \(p\).

mu

the mean parameter \(\mu\).

phi

the dispersion parameter \(\phi\).

lower.tail

logical; if TRUE (the default), computes \(F(y)=\Pr(Y\le y)\); if FALSE, computes the upper tail \(S(y)=\Pr(Y>y)\) directly, rather than as 1 - ptweedie_series(...).

log.p

logical; if TRUE, the logarithm of the probability is returned, computed directly via log-sum-exp accumulation over the series terms rather than as log(ptweedie_series(...)). This retains accuracy even when the probability itself would underflow to zero.

verbose

logical; if TRUE, displays some internal computation details. The default is FALSE.

details

logical; if TRUE, returns the value of the distribution function and some details.

max.terms

integer; the maximum number of terms summed for each value before the summing is truncated, and a warning produced. The default is 1e6, enough for \(\lambda\) up to about \(4\times 10^9\); each extra million terms takes about half a second per value.

Accuracy

The terms of the series are all positive, and are summed on the log scale, so both tails keep full relative accuracy, far into the tails. The only limitation is the number of terms: if more than max.terms are needed (when \(\lambda\) is very large), the series is truncated and a warning is given.

References

Dunn, Peter K and Smyth, Gordon K (Submitted). Evaluation of the Tweedie exponential dispersion model distribution function. Submitted.

Dunn, Peter K and Smyth, Gordon K (2005). Series evaluation of Tweedie exponential dispersion model densities Statistics and Computing, 15(4). 267--280. tools:::Rd_expr_doi("10.1007/s11222-005-4070-y")

Examples

Run this code
# Plot a Tweedie distribution function
y <- seq(0.01, 4, length = 50)
Fy <- ptweedie_series(y, power = 1.1, mu = 1, phi = 1)
plot(y, Fy, type = "l", lwd = 2, ylab = "Distribution function")

Run the code above in your browser using DataLab