lm.moments2: Fit normal linear model using pre-built matrix of second...
In gtx: Genetics ToolboX

Description Usage Arguments Details Value Author(s) Examples

Regression coefficients for a normal linear model are be calculated, using a pre-built sufficient summary statistic matrix that contains the second moments between all the variables of (potential) interest.

1	lm.moments2(xtx, leftvar, rightvars, n = NULL)

`xtx`	a matrix of second moments, typically built using `make.moments2`.
`leftvar`	name of the response variable (the left hand side of the formula).
`rightvars`	name(s) of the explanatory variables (the right hand side of the formula).
`n`	sample size, only needed if there is not a single intercept for all individuals.

Non-identifiable variables in rightvars are removed, with preference for keeping variables that occur earlier rather than later in rightvars.

A list with slots for the effect size estimates, standard errors, and a precision matrix.

This function is just a wrapper for calling est.moments2 with scale = NULL.

Toby Johnson Toby.x.Johnson@gsk.com

data(mthfrex)
xtx <- make.moments2(mthfr.params, c("SBP", "DBP", "SexC", "Age"), mthfrex)
lm.moments2(xtx, "SBP", c("ONE", "rs6668659_T", "rs4846049_T",
                                "rs1801133_G", "SexC", "Age"))
## Compare against results from lm
## Note have to use coded alleles in original data
lm(SBP ~ rs6668659_G+rs4846049_G+rs1801133_A+Sex+Age, data = mthfrex$data)
## Note in results Sex factor coded differently than SexC