timeLagLasso.path: Fit a path of time-lasso models

In orderedLasso: Ordered Lasso and Time-Lag Sparse Regression

Description

Fit a path of time-lasso models

Usage

timeLagLasso.path(x, y, lamlist = NULL, minlam = NULL, maxlam = NULL,
  nlam = 10, flmin = 0.01, strongly.ordered = TRUE, flmax = 1, maxlag,
  intercept = TRUE, standardize = TRUE, method = c("Solve.QP", "GG"),
  maxiter = 500, inneriter = 100, iter.gg = 100, trace = FALSE,
  epsilon = 1e-05)

Arguments

x	A matrix of predictors, where the rows are the samples and the columns are the predictors
y	A vector of observations, where length(y) equals nrow(x)
lamlist	Optional vector of values of lambda (the regularization parameter)
minlam	Optional minimum value for lambda
maxlam	Optional maximum value for lambda
nlam	Number of values of lambda to be tried
flmin	Fraction of maxlam minlam= flmin*maxlam. If computation is slow, try increasing flmin to focus on the sparser part of the path; default = 1e-2.
strongly.ordered	An option which allows users to order the coefficients in absolute value.
flmax	Multiplication of maxlam maxlam = flmax * maxlam. Default = 1
maxlag	Maximum time-lag chosen by user.
intercept	True if there is an intercept in the model.
standardize	Standardize the data matrix x.
method	Two options available, Solve.QP and Generalized Gradient
maxiter	Maximum iterations run by time-lag lasso. Initiazlied to 500.
inneriter	maximum iterations run by orderedLasso. Initialized to 100.
iter.gg	Maximum iterations run by generalized gradient. Intialized to 100
trace	Output option; trace = TRUE gives verbose output.
epsilon	Error tolerance parameter for convergence criterion; default 1e-5

Value

bp	p by nlam matrix of estimated positive coefficients(p=#variables)
bn	p by nlam matrix of estimated negative coefficients
beta	p by nlam matrix of estimated coefficients
b0	a vector of length nlam of estimated intercept
lamlist	Vector of values of lambda used
err	Vector of errors
maxlag	Maximum time-lag variable
p	The number of predictors
fited	a length(y) by nlam matrix of fitted values
call	The call to "timeLagLasso.path"

Examples

set.seed(3)
n = 50
maxlag = 5
num_rows_needed = n + maxlag + 1
sigma = 4
x = matrix(rnorm(num_rows_needed * 4), nrow = num_rows_needed)
x_new = time_lag_matrix(x, maxlag)
b = c(3,1,1,0,0,
      4,1,0,0,0,
      3,2,1,0,0,
      1,0,0,0,0)
y = x_new %*% b + sigma* rnorm(nrow(x_new))
y = as.vector(y)
y = c(y, rnorm(maxlag + 1))
path1 = timeLagLasso.path(x= x, y = y, maxlag = 5, method = "Solve.QP", strongly.ordered = TRUE)
plot(path1)

Example output

Loading required package: Matrix

orderedLasso documentation built on May 2, 2019, 6:36 a.m.