ls-means function

Compute LS-means (aka population means or marginal means)

LS-means (least squares means, also known as population means and as marginal means) for a range of model types.

LSmeans(object, effect = NULL, at = NULL, level = 0.95, ...)

## Default S3 method:
LSmeans(object, effect = NULL, at = NULL, level = 0.95, ...)

## S3 method for class 'lmerMod'
LSmeans(object, effect = NULL, at = NULL, level = 0.95, adjust.df = TRUE, ...)

popMeans(object, effect = NULL, at = NULL, level = 0.95, ...)

## Default S3 method:
popMeans(object, effect = NULL, at = NULL, level = 0.95, ...)

## S3 method for class 'lmerMod'
popMeans(object, effect = NULL, at = NULL, level = 0.95, adjust.df = TRUE, ...)

Arguments

• object: Model object
• effect: A vector of variables. For each configuration of these the estimate will be calculated.
• at: A list of values of covariates (including levels of some factors) to be used in the calculations
• level: The level of the (asymptotic) confidence interval.
• ...: Additional arguments; currently not used.
• adjust.df: Should denominator degrees of freedom be adjusted?

Returns

A dataframe with results from computing the contrasts.

Details

There are restrictions on the formulas allowed in the model object. For example having y ~ log(x) will cause an error. Instead one must define the variable logx = log(x) and do y ~ logx.

Note

LSmeans and popMeans are synonymous.

Warning

Notice that LSmeans and LE_matrix

fails if the model formula contains an offset (as one would have in connection with e.g. Poisson regression.

Examples

## Two way anova:

data(warpbreaks)

m0 <- lm(breaks ~ wool + tension, data=warpbreaks)
m1 <- lm(breaks ~ wool * tension, data=warpbreaks)
LSmeans(m0)
LSmeans(m1)

## same as:
K <- LE_matrix(m0);K
linest(m0, K)
K <- LE_matrix(m1);K
linest(m1, K)

LE_matrix(m0, effect="wool")
LSmeans(m0, effect="wool")

LE_matrix(m1, effect="wool")
LSmeans(m1, effect="wool")

LE_matrix(m0, effect=c("wool", "tension"))
LSmeans(m0, effect=c("wool", "tension"))

LE_matrix(m1, effect=c("wool", "tension"))
LSmeans(m1, effect=c("wool", "tension"))

## Regression; two parallel regression lines:

data(Puromycin)

m0 <- lm(rate ~ state + log(conc), data=Puromycin)
## Can not use LSmeans / LE_matrix here because of
## the log-transformation. Instead we must do:
Puromycin$lconc <- log( Puromycin$conc )
m1 <- lm(rate ~ state + lconc, data=Puromycin)

LE_matrix(m1)
LSmeans(m1)

LE_matrix(m1, effect="state")
LSmeans(m1, effect="state")

LE_matrix(m1, effect="state", at=list(lconc=3))
LSmeans(m1, effect="state", at=list(lconc=3))

## Non estimable contrasts

## ## Make balanced dataset
dat.bal <- expand.grid(list(AA=factor(1:2), BB=factor(1:3),
                            CC=factor(1:3)))
dat.bal$y <- rnorm(nrow(dat.bal))

## ## Make unbalanced dataset
#      'BB' is nested within 'CC' so BB=1 is only found when CC=1
#       and BB=2,3 are found in each CC=2,3,4
dat.nst <- dat.bal
dat.nst$CC <-factor(c(1, 1, 2, 2, 2, 2, 1, 1, 3, 3,
                      3, 3, 1, 1, 4, 4, 4, 4))

mod.bal  <- lm(y ~ AA + BB * CC,    data=dat.bal)
mod.nst  <- lm(y ~ AA + BB : CC, data=dat.nst)

LSmeans(mod.bal, effect=c("BB", "CC"))
LSmeans(mod.nst, effect=c("BB", "CC"))
LSmeans(mod.nst, at=list(BB=1, CC=1))

LSmeans(mod.nst, at=list(BB=1, CC=2))
## Above: NA's are correct; not an estimable function

if( require( lme4 )){
 warp.mm <- lmer(breaks ~ -1 + tension + (1|wool), data=warpbreaks)
 LSmeans(warp.mm, effect="tension")
 class(warp.mm)
 fixef(warp.mm)
 coef(summary(warp.mm))
 vcov(warp.mm)
 if (require(pbkrtest))
   vcovAdj(warp.mm)
}

LSmeans(warp.mm, effect="tension")

See Also

LE_matrix, linest

Author(s)

Søren Højsgaard, sorenh@math.aau.dk