nested: nested weighted means for experimental units

In byandell/pda: Practial Data Analysis for Designed Experiments

Description

This is under development. It is a simple approximation to reducemeasurements to a single value for each sized experimental unit.

Usage

nested( object, ...)
## Default S3 method:
nested(object, data, project = proj(object), ...)
## S3 method for class 'aovprojlist'
nested( object, data, factors, response="y",
  nesting=c(1,ncol(ref)), ref = projref( object ), ...)

Arguments

object	linear model fit object with Error() nesting or projection list
data	data frame used in object
project	projection of data using object
factors	vector of names of factors in object
response	name of response
nesting	levels of nesting to be averaged (1=intercept, 2=whole plot, 3=sub-plot,etc.)
ref	data frame of reference values from projection
...	additional arguments for projwt

Details

See nested.default and nested.aovprojlist for code details.

Value

Data frame with average response for unique factor combinations and additional column weight with counts of cases used in averages.

See Also

proj.

Examples

## Not run: 
nest.data <- nested( proj(object), data, c("A","b"), "y", c(1,2))
aov(y~A*B, nest.data)

## End(Not run)

byandell/pda documentation built on May 13, 2019, 9:27 a.m.