regressResiduals.Rd
Runs quantile regression on residuals of the model (calculates spaces around jstar quantile)
regressResiduals( reg_spec_data, ehat, ind_hat, tau, trunc, small, weights, algorithm = "rq.fit.sfn_start_val", ... )
reg_spec_data | result of ensureSpecRank function; regression matrix with full rank |
---|---|
ehat | current residuals; subset of which to be used as dependent column |
ind_hat | column vector indicating which rows to be used in quantile regression |
tau | estimated quantile |
trunc | Boolean value; if true, replace those dependent values less than small with small itself; else, only use rows with residuals greater than small |
small | Value used with trunc; values less than small 'blow up' too greatly when logged |
weights | vector of optional weights |
algorithm | The name of a function which will estimate a quantile regression.
Defaults to rq.fit.sfn_start_val. Must be a string, as it is passed to |
... | other arguments to the function specified by the algorithm argument |
List of estimated coefficients, warnings, iterations, and controls as in standard quantile regression function