Zero inflated Gamma regression {Rfast2}

Zero inflated Gamma regression

Description

Zero inflated Gamma regression.

Usage

zigamma.reg(y, x, full = FALSE, tol = 1e-07, maxiters = 100) 

Arguments

y

The dependent variable, a numerical vector with numbers, zeros and higher.

x

A numerical matrix with the indendent variables. We add, internally, the first column of ones.

full

If this is FALSE, the coefficients and the log-likelihood will be returned only. If this is TRUE, more information is returned.

tol

The tolerance value to terminate the Newton-Raphson algorithm.

maxiters

The maximum number of iterations that can take place in each regression.

Details

Two regression models are fitted, a binary logistic regression and a Gamma regression model to the non-zero responses.

Value

Depending on whether "full" is TRUE or not different outputs are returned. In general, the regression coefficients, the iterations required by Newton-Raphson and the deviances are returned. If full is TRUE, a matrix with their standard errors and the Wald test statistics is returned as well.

Author(s)

Michail Tsagris.

R implementation and documentation: Michail Tsagris [e-mail].

References

Mills, Elizabeth Dastrup (2013). Adjusting for covariates in zero-inflated gamma and zero-inflated log-normal models for semicontinuous data. PhD thesis, University of Iowa.

See Also

zigamma.mle, ztp.reg

Examples

y <- rgamma(100, 4, 1)
y[sample(100, 10)] <- 0
x <- rnorm(100)
a <- zigamma.reg(y, x)