# SattDF: Satterthwaite Approximation for Total Degrees of Freedom and... In VCA: Variance Component Analysis

 SattDF R Documentation

## Satterthwaite Approximation for Total Degrees of Freedom and for Single Variance Components

### Description

This function estimates degrees of freedom of the total variance (type="total") in random models or individual variance components (type="individual"). It bases on the results of the unified approach to ANOVA-type estimation of variance components as implemented in functions `anovaVCA` and `anovaMM`.

### Usage

``````SattDF(MS, Ci, DF, type = c("total", "individual"))
``````

### Arguments

 `MS` (numeric) vector of sequential mean squares (ANOVA type-1). `Ci` (matrix) where elements are numeric values representing the inverse of the coefficient matrix for calculation of expected mean squares (see `anovaVCA`). `DF` (numeric) vector with the degrees of freedom for each factor in a ANOVA type-1 model. `type` (character) string specifying whether "total" degrees of freedom should be approximated or those of individual variance components

### Details

Function is used internally, thus, it is not exported. Option 'type="total"' is used in functions `anovaVCA` and `anovaMM` for approximating total DF. Option 'type="individual"' is used in function `VCAinference` when choosing 'ci.method="satterthwaite"' for approximating DFs for individual variance components.

### Value

numeric value representing the Satterthwaite DFs of the total variance.

### Author(s)

Andre Schuetzenmeister andre.schuetzenmeister@roche.com

### Examples

``````
## Not run:
data(dataEP05A2_2)
res <- anovaVCA(y~day/run, dataEP05A2_2)
VCA:::SattDF(res\$aov.tab[-1,"MS"], getMat(res, "Ci.MS"), res\$aov.tab[-1,"DF"], type="tot")

# now approximating individual DF for variance components
VCA:::SattDF(res\$aov.tab[-1,"MS"], getMat(res, "Ci.MS"), res\$aov.tab[-1,"DF"], type="i")

## End(Not run)
``````

VCA documentation built on May 29, 2024, 1:48 a.m.