Fit Linear Regression For Many Responses

Fit regression model Y[j,] ~ design for each response j

Usage

# S4 method for class 'sparseMatrix'
lmFitResponses(
  Y,
  design,
  Weights,
  detail = 1,
  lambda = 0,
  nthreads = 1,
  ...,
  chunkSize = 1000,
  verbose = TRUE
)

# S4 method for class 'ANY'
lmFitResponses(
  Y,
  design,
  Weights,
  detail = 1,
  lambda = 0,
  nthreads = 1,
  ...,
  chunkSize = 1000,
  verbose = TRUE
)

# S4 method for class 'DelayedArray'
lmFitResponses(
  Y,
  design,
  Weights,
  detail = 1,
  lambda = 0,
  nthreads = 1,
  ...,
  chunkSize = 1000,
  verbose = TRUE
)

Arguments

Y

matrix of responses as __rows__

design

design matrix

Weights

matrix sample-level weights the same dimension as Y

detail

level of model detail returned, with LOW = 0, MEDIUM = 1, HIGH = 2. LOW (beta, se, sigSq, rdf), MEDIUM (vcov), HIGH (residuals), MOST (hatvalues)

lambda

ridge shrinkage parameter

nthreads

number of threads. Each model is fit in serial, analysis is parallelized across responses.

...

other args

chunkSize

number of features to read per chunk

verbose

show progress

Value

List of parameter estimates with entries coef, se, sigSq, rdf and other depending on detail

Details

Since the weights vary for each response, each model is computed separately without recycling precomputed values

Examples

library(DelayedArray)
n <- 100
m <- 5
nc <- 2
set.seed(1)
Y <- matrix(rnorm(n * m), m, n)
Y <- DelayedArray(Y)
X <- matrix(rnorm(n * nc), n, nc)
rownames(Y) <- seq(m)
W <- matrix(runif(n * m), m, n)

# fit regressions with model j using Y[j,] as a response
fit <- lmFitResponses(Y, X, W)

fit
#> 		 lmFitResponses 
#> 
#> coefs(2): V1, V2
#> responses(5): 1, 2, 3, 4, 5
#> family: gaussian/identity 
#> Estimated: se, dispersion, rdf 
#>