kclass.fit: Low-level kclass instrumental variables function.
In potterzot/rkclass: K-class regressions (Two-Stage Least Squares, LIML, Fuller)

Low-level kclass instrumental variables function.

1
2
3

kclass.fit(x, y, z = NULL, weights = NULL, offset = NULL, k = NULL,
  alpha = NULL, model.type = c("TSLS", "LIML", "FULLER", "KCLASS", "OLS"),
  eig = c("eigen", "geigen"), na.action, ...)

`x`	data frame or matrix of endogenous regressors.
`y`	data frame or vector of the dependent variable.
`z`	data frame or matrix of instruments and/or exogenous regressors.
`weights`	sample weights. Default is NULL.
`offset`	model offset.
`k`	the k parameter value. If k is set, `model.type` is set to "KCLASS".
`alpha`	the fuller parameter, only used if `model.type` is "FULLER".
`model.type`	string determining model type. Must be one of "OLS", "LIML", "TSLS", "KCLASS", "FULLER". Default is LIML."
`eig`	eigenvalue method. One of "eigen" or "geigen".
`na.action`	function indicating what to do when data contain NAs. The default is set by the na.action setting of options
`...`	additional parameters not currently used.