mpdi_obj: objective function for probability of a prediciton interval...

In femiguez/predintma: Prediction Intervals for Meta-Analysis

Description

objective function for probability of a prediciton interval for 'small' sample sizes.

Usage

mpdi_obj(
  alpha,
  x,
  method = c("tdist", "conformal"),
  m.method = c("quantile", "deviation", "jackknife"),
  alpha.penalty = 0,
  scale = FALSE
)

Arguments

alpha	miscoverage or 'error rate'
x	a vector
method	either 'tdist' (assumes normality) or 'conformal' (distribution-free)
m.method	method used to compute conformal prediction interval: either "quantile", "deviation" or "jackknife"
alpha.penalty	whether to include an alpha penalty (default 0 or 'no')
scale	whether to scale the input vector. This only makes sense if the alpha.penalty is different from zero.

Details

Objective funciton for optimization

The idea is to find the maximum level of probability that will produce a prediction interval which matches the minimum and maximum values in the observed sample. The distance is calculated as abs(x.min - calc.lower.bound) + abs(x.max - calc.upper.bound).

Value

a single value which represents a value that should be minimized

Examples

## Not run: 
set.seed(12345)
x <- rnorm(10)
alphas <- seq(0,1, 0.05)
objf <- numeric(length(alphas))
for(i in 1:length(objf)){
  objf[i] <- mpdi_obj(alphas[i], x = x, method = "conformal")
}
qplot(alphas, objf, geom = "line")
## Trying the t-distribution
y <- rt(10, df = 1)
alphas <- seq(0,1, 0.05)
objf <- numeric(length(alphas))
for(i in 1:length(objf)){
  objf[i] <- mpdi_obj(alphas[i], x = y, method = "conformal")
}
qplot(alphas, objf, geom = "line")

## End(Not run)

femiguez/predintma documentation built on July 5, 2021, 4:16 a.m.