mcpfisher: Fisher's Protected Least Significance Difference

In joemckean/mathstat: Introduction to Mathematical Statistics R Package

Description

Fisher's PLSD Multiple Comparison Procedure is a two-stage procedure. For a specified level of significance (alpha), the first step consists of the F-test of the hypotheses of equal means. If the test is rejected at level alpha, then the second stage will compute pairwise (1-alpha)100% confidence intervals.

Usage

mcpfisher(y, ind, alpha = 0.05)

Arguments

y	Vector of the combined samples
ind	Corresponding treatment vector
alpha	Level of significance used for Fisher's Procedure

Details

An example of this Fisher's PLSD Multiple Comparison Procedure can be found in Example 9.4.1 on page 529 of HMC (2018).

Value

This function returns a table to the user. The first two columns j & jp are the specific levels (categories of cars) being compared. muj & mujp are the means of two levels being compared. diff is the difference in means between these levels. se is the standard error for each comparison, err is the comparison's error estimate, and lb & ub consist of the lower and upper bound for each comparison's confidence interval.

References

Hogg, R., McKean, J., Craig, A. (2018) Introduction to Mathematical Statistics, 8th Ed. Boston: Pearson.

See Also

mcpbon() for details Bonferroni procedure for all pairwise comparisons

Examples

speed <- fastcars$speed
car <- fastcars$ind
mcpfisher(speed, car)

joemckean/mathstat documentation built on May 30, 2019, 2:01 p.m.