LL1WayAnova: Log-likelihood ratio statistic for grouped responses

In EffectsRelBaseline: Test changes of a grouped response relative to baseline.

Description

Computes the log-likelihood ratio statistic for a test of the null hypothesis that the responses are all drawn from one group, rather than the alternative hypothesis that each group has a separate mean.

Usage

LL1WayAnova(resp, groups)

Arguments

resp	response values
groups	grouping variables (will be treated as factors)

Value

log-likelihood ratio

Note

This is the normal likelihood-ratio which is used in performing a 1-way ANOVA. It is included here for comparison.

Author(s)

Peter N. Steinmetz <[e-mail]>

References

B. Lindgren, Statistical Theory, 4th Ed. (1993), section 14.7.

See Also

fRatioStat,relBackgroundLL,sumSqCat

Examples

data(ef)
LL1WayAnova(ef$absResp,ef$race)

Example output

[1] -8.846341

EffectsRelBaseline documentation built on May 2, 2019, 9:44 a.m.