lace: Localised average causal effects
In jrs95/nlmr: Non-linear Mendelian randomisation

lace	R Documentation

Localised average causal effects

lace computes the localised average causal effect for each quantile.

lace(
  y,
  x,
  g,
  covar = NULL,
  q,
  x0q,
  xc_sub = TRUE,
  family = "gaussian",
  xpos = "mean"
)

`y`	vector of outcome values
`x`	vector of exposure values
`g`	the instrumental variable
`covar`	`data.frame` of covariates
`q`	the number of quantiles the exposure distribution is to be split into within which a causal effect will be fitted, known as localised average causal effects (LACE) (default: `10`)
`x0q`	quantiles of x0 (the IV-free exposure)
`xc_sub`	compute the association between the exposure and the insturment in each quantile of x0
`family`	a description of the error distribution and link function to be used in the model and is a `character` string naming either the gaussian (i.e. `"gaussian"` for continuous data) or binomial (i.e. `"binomial"` for binary data) family function (default: `"gaussian"`)
`xpos`	the position used to relate `x` to the localised average causal effect, this can either be the mean of the x-values within each quantile or a percentile (e.g. 0.5 corresponds to the median value) (default: `"mean"`)

lace returns a data.frame of the LACE for quantile:

`coef`	the LACE in each quantile
`coef_se`	the standard error of the LACE in each quantile
`xmean`	the mean of the exposure in each quantile
`xcoef_sub`	the association between the exposure and the instrument in each quantile
`xcoef_sub_se`	the standard error of the association between the exposure and the instrument in each quantile