tolerance (version 0.4.0)

K.table: Tables of K-factors for Tolerance Intervals Based on Normality

Description

Tabulated summary of k-factors for tolerance intervals based on normality. The user can specify multiple values for each of the three inputs.

Usage

K.table(n, alpha, P, side = 1, f = NULL,
        by.arg = c("n", "alpha", "P"))

Arguments

n

A vector of (effective) sample sizes.

alpha

The level chosen such that 1-alpha is the confidence level. Can be a vector.

P

The proportion of the population to be covered by this tolerance interval. Can be a vector.

side

Whether a 1-sided or 2-sided tolerance interval is required (determined by side = 1 or side = 2, respectively).

f

The number of degrees of freedom associated with calculating the estimate of the population standard deviation. If NULL, then f is taken to be n-1. Only a single value can be specified for f.

by.arg

How you would like the output organized. If by.arg = "n", then the output provides a list of matrices sorted by the values specified in n. The matrices have rows corresponding to the values specified by 1-alpha

Value

• K.table returns a list with a structure determined by the argument by.arg described above.

Details

The method used for estimating the k-factors is that due to Howe as it is generally viewed as more accurate than the Weissberg-Beatty method.

References

Howe, W. G. (1969), Two-Sided Tolerance Limits for Normal Populations - Some Improvements, Journal of the American Statistical Association, 64, 610--620. Weissberg, A. and Beatty, G. (1969), Tables of Tolerance Limit Factors for Normal Distributions, Technometrics, 2, 483--500.

See Also

K.factor

Examples

## Tables generated for each value of the sample size.

K.table(n = seq(50, 100, 10), alpha = c(0.01, 0.05, 0.10), 
        P = c(0.90, 0.95, 0.99), by.arg = "n")

## Tables generated for each value of the confidence level.

K.table(n = seq(50, 100, 10), alpha = c(0.01, 0.05, 0.10), 
        P = c(0.90, 0.95, 0.99), by.arg = "alpha")

## Tables generated for each value of the coverage proportion.

K.table(n = seq(50, 100, 10), alpha = c(0.01, 0.05, 0.10), 
        P = c(0.90, 0.95, 0.99), by.arg = "P")