jmdem (version 1.0.1)

summary.jmdem.sim: Summarising JMDEM Simulations

Description

These functions are all methods for class jmdem.sim or summary.jmdem.sim objects.

Usage

# S3 method for jmdem.sim
summary(object, digits = max(3L, getOption("digits") - 3L), 
        scientific = FALSE, pvalue = 0.05, 
        minv.method = c("solve", "chol2inv", "ginv"), 
        other.call = FALSE, details = FALSE, ...)
              
# S3 method for summary.jmdem.sim
print(x, digits = max(3L, getOption("digits") - 3L), scientific = FALSE, 
      pvalue = 0.05, signif.stars = getOption("show.signif.stars"), 
      other.call = FALSE, details = FALSE, ...)

Arguments

object

an object of class "jmdem.sim", usually, a result of a call to jmdem.sim.

x

an object of class "summary.jmdem.sim", usually, a result of a call to summary.jmdem.sim.

digits

the number of significant digits to use when printing.

scientific

logical; if TRUE, scientific notation is used when printing.

pvalue

a value between 0 and 1. It is used to compute the coverage proportion of the true parameter values by the simulated fits.

minv.method

the method used to invert matrices during the estimation process. "solve" gives the solutions of a system of equations, "chol2inv" gives the inverse from Choleski or QR decomposition and "ginv" gives the generalised inverse of a matrix. If none of the methods is specified or if they are specified in a vector such as c("solve", "chol2inv", "ginv"), the matrices will be inverted by the methods in the sequence as given in the vector until it is found.

signif.stars

logical. If TRUE, 'significance stars' are printed for each coefficient.

other.call

logical. If true, the rest of simulation call (i.e. without the mean and dispersion submodel formulas, families, true values) will be shown.

details

logical. If true, coefficients, standard errors, true parameter coverage (TRUE/FALSE) and asymptotic test statistics of each simulation will be listed.

...

further arguments passed to or from other methods.

Value

digits

the number of significant digits to use when printing.

scientific

logical value of using scientific notation when printing.

details

logical value of printing details of each simulation.

other.call

logical value of printing other parameters of the simulation call.

pvalue

numeric value between 0 and 1 used for the computation of the true parameter coverage.

beta.true

user-defined vector containing the true parameter values of the mean submodel.

lambda.true

user-defined vector containing the true parameter values of the dispersion submodel.

simcall

the component from object.

mformula

the component from object.

dformula

the component from object.

mfamily

the component from object.

dfamily

the component from object.

coefficients

mean and dispersion submodel parameter coefficients fitted in each simulation saved in a data.frame.

stderr

standard erros of all mean and dispersion submodel parameter coefficients estimated in each simulation saved in a data.frame.

iterations

a vectror containing the running numbers of each simulation.

confint

confidence intervals of all mean and dispersion submodel parameter coefficients estimated in each simulation saved in a data.frame.

coverage

the coverage of all true submodel parameters by the confidence intervals estimated in each simulation saved in a data.frame.

asymp.test

Rao's score and Wald test statistics of each simulation saved in a data.frame.

average.summary

Arithmetric means of the coefficients, standard errors, confidence interval converage estimated in all simulations saved in a data.frame.

average.asymp.test

(Arithmetric means of the Rao's score and Wald test statistics estimated in all simulations saved in a data.frame.

Details

The arithmetric mean of the coefficients, standard errors and coverage by the confidence intervals estimated in all simulations will be listed in a table. A detail listing of each simulation's results can be provided if required by details = TRUE. The summary also includes the averages of the Rao's score and Wald test statistics of all simulation fits.

print.summary.jmdem.sim tries to be smart about formatting the coefficients, standard errors, etc according the number of significant digits (default of user-specified) or the usage of scientific notation or not.

See Also

jmdem.sim, jmdem, summary

Examples

# NOT RUN {
## Example in jmdem.sim(...)
sim <- jmdem.sim(mformula = y ~ x, dformula = ~ z, beta.first = TRUE, 
                 mfamily = gaussian, dfamily = Gamma(link = "log"), 
                 x.str = list(type = "numeric", random.func = "rnorm", 
                              param = list(mean = 0, sd = 2)),
                 z.str = list(type = "factor", random.func = "runif", 
                              param = list(min = 0, max = 2)),
                 beta.true = c(1.5, 4), lambda.true = c(2.5, 3, -0.2), 
                 grad.func = TRUE, method = "BFGS", n = 50,
                 simnum = 10)
                  
## Summarise simulation
summary(sim, details = FALSE, other.call = TRUE)
# }