ldf {flashier}R Documentation

LDF method for flash and flash fit objects

Description

Given a flash or flash_fit object, returns the LDF decomposition YLDFY \approx LDF'.

Usage

ldf(object, type)

## S3 method for class 'flash'
ldf(object, type = "f")

## S3 method for class 'flash_fit'
ldf(object, type = "f")

Arguments

object

An object inheriting from class flash or flash_fit.

type

Takes identical arguments to function norm. Use "f" or "2" for the 2-norm (Euclidean norm); "o" or "1" for the 1-norm (taxicab norm); and "i" or "m" for the infinity norm (maximum norm).

Details

When the prior families G(k)G_\ell^{(k)} and Gf(k)G_f^{(k)} are closed under scaling (as is typically the case), then the EBMF model (as described in the documention to function flash) is only identifiable up to scaling by a diagonal matrix DD:

Y=LDF+E.Y = LDF' + E.

Method ldf scales columns k\ell_k and fkf_k so that, depending on the argument to parameter type, their 1-norms, 2-norms, or infinity norms are equal to 1.

Value

A list with fields L, D, and F, each of which corresponds to one of the matrices in the decomposition YLDFY \approx LDF' (with the columns of LL and FF scaled according to argument type). Note that D is returned as a vector rather than a matrix (the vector of diagonal entries in DD). Thus, "fitted values" LDFLDF' can be recovered as L %*% diag(D) %*% t(F).

Methods (by class)


[Package flashier version 1.0.7 Index]