estimable {gregmisc} | R Documentation |
Compute and test estimable linear functions of the fitted coefficients (including contrasts) of regression objects
estimable(obj, cm, conf.int=NULL)
obj |
Regression (lm,glm,lme) object. |
cm |
matrix specifying estimable linear functions or contrasts (one per row). The number of columns must match the number of fitted coefficients in the model. |
conf.int |
Confidence level. If provided, confidence intervals will be computed. |
Computes an estimate, test statitic, significance test, and (optional)
confidence interval for each linear functions of the model
coefficients specified by the rows of cm
. The estimates and
their variances are obtained by applying the matrix cm
to the
model estimates variance-covariance matrix. Degrees of freedom are
obtained from the appropriate model terms.
The user is responsible for ensuring that the specified
linear functions are meaningful. For computing contrasts among levels
of a single factor, contrast.lm
may be more
convenient.
Returns a matrix with one row per linear function. Columns contain estimated coefficients, standard errors, t values, degrees of freedom, two-sided p-values, and the lower and upper endpoints of the 1-alpha confidence intervals.
The estimated fixed effect parameter of lme
objects may have
different degrees of freedom. If a specified contrast includes
nonzero coefficients for parameters with differing degrees of freedom,
the smallest number of degrees of freedom is used and a warning
message is issued.
BXC (Bendix Carstensen) bxc@novonordisk.com and Gregory R. Warnes Gregory_R_Warnes@groton.pfizer.com
contrast.lm
,
lm
, lme
,
contrasts
,
contr.treatment
, contr.poly
y <- rnorm(100) x <- cut(rnorm(100, mean=y, sd=0.25),c(-4,-1.5,0,1.5,4)) reg <- lm(y ~ x) summary(reg) # look at the group means gm <- sapply(split(y,x),mean) gm # contrast mean of 2nd group vs mean of 4th group estimable(reg, c( 0, 1, 0, -1) ) # estimate should be equal to: gm[2] - gm[4] # confidence intervals etc. for the line for level 4 # for a separate continuous variable modelled as spline # with a single knot at 0.5: x2 <- rnorm(100,mean=y,sd=0.5) reg2 <- lm(y ~ x + x2 + pmax(x2-0.5,0) ) xx2<-seq(-2,2,,50) tmp <- estimable(reg2,cbind(1,0,0,1,xx2,pmax(xx2-0.5,0)), conf.int=0.95) plotCI(x=xx2,y=tmp[,1],li=tmp[,6],ui=tmp[,7])