oomlm_robust: Out of memory Linear model with robust standard errors

In blakeboswell/ploom: Out-of-memory and memory efficient Linear Model fitting with support for Inference

Description

Perform memory-efficient generalized linear regression using the AS274 bounded memory QR factorization algorithm and estimate robust standard errors.

Usage

oomlm_robust(formula, weights = NULL, se_type = "HC1", ...)

Arguments

formula	A symbolic description of the model to be fitted of class formula.
weights	A one-sided, single term formula specifying weights.
se_type	Indicates what Standard Error method to use: "HC0", "HC1" "stata", or "classical".
...	Ignored.

Details

The provided formula must not contain any data-dependent terms to ensure consistency across calls to fit(). Factors are permitted, but the levels of the factor must be the same across all data chunks. Empty factor levels are accepted.

Value

A oomlm_robust model is perpetually in an in-progress state. It is up to the user to know when fitting is complete. Therefore, only basic model characteristics are provided as values. Statistics are available on demand via summary and extractor functions.

n	The number of observations processed.
df.residual	The residual degrees of freedom.
formula	The formula object specifying the linear model.
terms	The terms object specifying the terms of the linear model.
weights	a one-sided, single term formula specifying weights.
call	The matched call.

See Also

oomlm()

blakeboswell/ploom documentation built on May 25, 2019, 3:24 p.m.