Estimation of the parameters of a reciprocal inverse Gaussian distribution.

rigff(lmu = "identitylink", llambda = "loglink", imu = NULL,
      ilambda = 1)

Arguments

lmu, llambda

Link functions for mu and lambda. See Links for more choices.

imu, ilambda

Initial values for mu and lambda. A NULL means a value is computed internally.

Details

See Jorgensen (1997) for details.

Value

An object of class "vglmff" (see vglmff-class). The object is used by modelling functions such as vglm, and vgam.

References

Jorgensen, B. (1997). The Theory of Dispersion Models. London: Chapman & Hall

Author

T. W. Yee

Note

This distribution is potentially useful for dispersion modelling.

See also

Examples

rdata <- data.frame(y = rchisq(100, df = 14))  # Not 'proper' data!!
fit <- vglm(y ~ 1, rigff, rdata, trace = TRUE)
#> Iteration 1: loglikelihood = -204.66436
#> Iteration 2: loglikelihood = -203.31288
#> Iteration 3: loglikelihood = -203.28874
#> Iteration 4: loglikelihood = -203.28873
#> Iteration 5: loglikelihood = -203.28873
fit <- vglm(y ~ 1, rigff, rdata, trace = TRUE, crit = "c")
#> Iteration 1: coefficients = 12.70634373, -0.32424612
#> Iteration 2: coefficients = 12.31959523, -0.47208282
#> Iteration 3: coefficients = 12.25397505, -0.49405914
#> Iteration 4: coefficients = 12.25250125, -0.49452809
#> Iteration 5: coefficients = 12.25250056, -0.49452831
#> Iteration 6: coefficients = 12.25250056, -0.49452831
summary(fit)
#> 
#> Call:
#> vglm(formula = y ~ 1, family = rigff, data = rdata, trace = TRUE, 
#>     crit = "c")
#> 
#> Coefficients: 
#>               Estimate Std. Error z value Pr(>|z|)    
#> (Intercept):1  12.2525     0.4482  27.335  < 2e-16 ***
#> (Intercept):2  -0.4945     0.1414  -3.497 0.000471 ***
#> ---
#> Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#> 
#> Names of linear predictors: mu, loglink(lambda)
#> 
#> Log-likelihood: -203.2887 on 198 degrees of freedom
#> 
#> Number of Fisher scoring iterations: 6 
#> 
#> No Hauck-Donner effect found in any of the estimates
#>