Summary Statistics for Final Cycle

Generates summary statistics of final population model parameters.

Usage

# S3 method for class 'PM_final'
summary(object, lower = 0.025, upper = 0.975, file = NULL, ...)

Arguments

object: The PM_final object made after an NPAG or IT2B run
lower: Desired lower confidence interval boundary. Default is 0.025. Ignored for IT2B objects.
upper: Desired upper confidence interval boundary. Default is 0.975. Ignored for IT2B objects.
file: Filename to save the summary. Include path if necessary.
...: Not used.

Value

The output is a data frame. For NPAG this has 4 columns:

value The value of the summary statistic
par The name of the parameter
type Either WtMed for weighted median, or MAWD for MAWD (see details)
percentile Requested lower, 0.5 (median), and upper quantiles For IT2B this has 6 columns:
mean Parameter mean value
se.mean Standard error of the mean
cv.mean Error of the mean divided by mean
var Variance of the parameter values
se.var Standard error of the variance
summary Name of the summary statistic

Details

#' This is a function usually called by the $summary() method for PM_final objects within a PM_result. The function can be called directly on a PM_final object. For NPAG runs, this function will generate weighted medians as central tendencies of the population points with a 95% confidence interval (95% CI) around the median, and the median absolute weighted deviation (MAWD) from the median as a measure of the variance, with its 95% CI. These estimates correspond to weighted mean, 95% CI of the mean, variance, and 95% CI of the variance, respectively, for a sample from a normal distribution.

To estimate these non-parametric summaries, the function uses a Monte Carlo simulation approach, creating 1000 x npoint samples with replacement from the weighted marginal distribution of each parameter, where npoint is the number of support points in the model. As an example, if there are 100 support points, npoint = 100, and for Ka, there will be 1000 sets of 100 samples drawn from the weighted marginal distribution of the values for Ka. For each of the 1,000 sets of npoint values, the median and MAWD are calculated, with MAWD equal to the median absolute difference between each point and the median of that set. The output is npoint estimates of the weighted median and npoint estimates of the MAWD for each parameter, from which the median, 2.5th, and 97.5th percentiles can be found as point estimates and 95% confidence interval limits, respectively, of both the weighted median and MAWD.

For IT2B runs, the function will return the mean and variance of each parameter, and the standard errors of these terms, using $$SE_mean = SD/\sqrt(nsub)$$ $$SE_var = var * \sqrt(2/(nsub-1))$$.

Author

Michael Neely

Examples

if (FALSE) { # \dontrun{
NPex$final$summary() # preferred
ITex$final$summary()
summary(NPex$final) # alternate
} # }