Fit a generalized linear model regularized with the sorted L1 norm, which applies a non-increasing regularization sequence to the coefficient vector (\(\beta\)) after having sorted it in decreasing order according to its absolute values.
SLOPE(
x,
y,
family = c("gaussian", "binomial", "multinomial", "poisson"),
intercept = TRUE,
center = !inherits(x, "sparseMatrix"),
scale = c("l2", "l1", "sd", "none"),
alpha = c("path", "estimate"),
lambda = c("bh", "gaussian", "oscar"),
alpha_min_ratio = if (NROW(x) < NCOL(x)) 0.01 else 1e-04,
path_length = if (alpha[1] == "estimate") 1 else 20,
q = 0.1 * min(1, NROW(x)/NCOL(x)),
screen = TRUE,
screen_alg = c("strong", "previous"),
tol_dev_change = 1e-05,
tol_dev_ratio = 0.995,
max_variables = NROW(x),
solver = c("fista", "admm"),
max_passes = 1e+06,
tol_abs = 1e-05,
tol_rel = 1e-04,
tol_rel_gap = 1e-05,
tol_infeas = 0.001,
tol_rel_coef_change = 0.001,
diagnostics = FALSE,
verbosity = 0,
sigma,
n_sigma,
lambda_min_ratio
)
the design matrix, which can be either a dense matrix of the standard matrix class, or a sparse matrix inheriting from Matrix::sparseMatrix. Data frames will be converted to matrices internally.
the response, which for family = "gaussian"
must be numeric; for
family = "binomial"
or family = "multinomial"
, it can be a factor.
model family (objective); see Families for details.
whether to fit an intercept
whether to center predictors or not by their mean. Defaults
to TRUE
if x
is dense and FALSE
otherwise.
type of scaling to apply to predictors.
"l1"
scales predictors to have L1 norms of one.
"l2"
scales predictors to have L2 norms of one.#'
"sd"
scales predictors to have a population standard deviation one.
"none"
applies no scaling.
scale for regularization path: either a decreasing numeric vector (possibly of length 1) or a character vector; in the latter case, the choices are:
"path"
, which computes a regularization sequence
where the first value corresponds to the intercept-only (null) model and
the last to the almost-saturated model, and
"estimate"
, which estimates a single alpha
using Algorithm 5 in Bogdan et al. (2015).
When a value is manually entered for alpha
, it will be scaled based
on the type of standardization that is applied to x
. For scale = "l2"
,
alpha
will be scaled by \(\sqrt n\). For scale = "sd"
or "none"
,
alpha will be scaled by \(n\), and for scale = "l1"
no scaling is
applied. Note, however, that the alpha
that is returned in the
resulting value is the unstandardized alpha.
either a character vector indicating the method used to construct the lambda path or a numeric non-decreasing vector with length equal to the number of coefficients in the model; see section Regularization sequences for details.
smallest value for lambda
as a fraction of
lambda_max
; used in the selection of alpha
when alpha = "path"
.
length of regularization path; note that the path
returned may still be shorter due to the early termination criteria
given by tol_dev_change
, tol_dev_ratio
, and max_variables
.
parameter controlling the shape of the lambda sequence, with
usage varying depending on the type of path used and has no effect
is a custom lambda
sequence is used.
whether to use predictor screening rules (rules that allow some predictors to be discarded prior to fitting), which improve speed greatly when the number of predictors is larger than the number of observations.
what type of screening algorithm to use.
"strong"
uses the set from the strong screening rule and check
against the full set
"previous"
first fits with the previous active set, then checks
against the strong set, and finally against the full set if there are
no violations in the strong set
the regularization path is stopped if the fractional change in deviance falls below this value; note that this is automatically set to 0 if a alpha is manually entered
the regularization path is stopped if the deviance ratio \(1 - \mathrm{deviance}/\mathrm{(null-deviance)}\) is above this threshold
criterion for stopping the path in terms of the maximum number of unique, nonzero coefficients in absolute value in model. For the multinomial family, this value will be multiplied internally with the number of levels of the response minus one.
type of solver use, either "fista"
or "admm"
;
all families currently support FISTA but only family = "gaussian"
supports ADMM.
maximum number of passes (outer iterations) for solver
absolute tolerance criterion for ADMM solver
relative tolerance criterion for ADMM solver
stopping criterion for the duality gap; used only with FISTA solver.
stopping criterion for the level of infeasibility; used with FISTA solver and KKT checks in screening algorithm.
relative tolerance criterion for change in coefficients between iterations, which is reached when the maximum absolute change in any coefficient divided by the maximum absolute coefficient size is less than this value.
whether to save diagnostics from the solver (timings and other values depending on type of solver)
level of verbosity for displaying output from the program. Setting this to 1 displays basic information on the path level, 2 a little bit more information on the path level, and 3 displays information from the solver.
deprecated; please use alpha
instead
deprecated; please use path_length
instead
deprecated; please use alpha_min_ratio
instead
An object of class "SLOPE"
with the following slots:
a three-dimensional array of the coefficients from the model fit, including the intercept if it was fit. There is one row for each coefficient, one column for each target (dependent variable), and one slice for each penalty.
a three-dimensional logical array indicating whether a coefficient was zero or not
the lambda vector that when multiplied by a value in alpha
gives the penalty vector at that point along the regularization
path
vector giving the (unstandardized) scaling of the lambda sequence
a character vector giving the names of the classes for binomial and multinomial families
the number of passes the solver took at each step on the path
the number of violations of the screening rule at each step on the path;
only available if diagnostics = TRUE
in the call to SLOPE()
.
a list where each element indicates the indices of the coefficients that were active at that point in the regularization path
the number of unique predictors (in absolute value)
the deviance ratio (as a fraction of 1)
the deviance of the null (intercept-only) model
the name of the family used in the model fit
a data.frame
of objective values for the primal and dual problems, as
well as a measure of the infeasibility, time, and iteration; only
available if diagnostics = TRUE
in the call to SLOPE()
.
the call used for fitting the model
Gaussian
The Gaussian model (Ordinary Least Squares) minimizes the following objective: $$ \frac{1}{2} \Vert y - X\beta\Vert_2^2 $$
Binomial
The binomial model (logistic regression) has the following objective: $$ \sum_{i=1}^n \log\left(1+ \exp\left(- y_i \left(x_i^T\beta + \beta_0 \right) \right) \right) $$ with \(y \in \{-1, 1\}\).
Poisson
In poisson regression, we use the following objective:
$$ -\sum_{i=1}^n \left(y_i\left(x_i^T\beta + \beta_0\right) - \exp\left(x_i^T\beta + \beta_0\right)\right) $$
Multinomial
In multinomial regression, we minimize the full-rank objective $$ -\sum_{i=1}^n\left( \sum_{k=1}^{m-1} y_{ik}(x_i^T\beta_k + \beta_{0,k}) - \log\sum_{k=1}^{m-1} \exp\big(x_i^T\beta_k + \beta_{0,k}\big) \right) $$ with \(y_{ik}\) being the element in a \(n\) by \((m-1)\) matrix, where \(m\) is the number of classes in the response.
There are multiple ways of specifying the lambda
sequence
in SLOPE()
. It is, first of all, possible to select the sequence manually by
using a non-increasing
numeric vector, possible of length one, as argument instead of a character.
If all lambda
are the same value, this will
lead to the ordinary lasso penalty. The greater the differences are between
consecutive values along the sequence, the more clustering behavior
will the model exhibit. Note, also, that the scale of the \(\lambda\)
vector makes no difference if alpha = NULL
, since alpha
will be
selected automatically to ensure that the model is completely sparse at the
beginning and almost unregularized at the end. If, however, both
alpha
and lambda
are manually specified, both of the scales will
matter.
Instead of choosing the sequence manually, one of the following automatically generated sequences may be chosen.
BH (Benjamini--Hochberg)
If lambda = "bh"
, the sequence used is that referred to
as \(\lambda^{(\mathrm{BH})}\) by Bogdan et al, which sets
\(\lambda\) according to
$$
\lambda_i = \Phi^{-1}(1 - iq/(2p)),
$$
for \(i=1,\dots,p\), where \(\Phi^{-1}\) is the quantile
function for the standard normal distribution and \(q\) is a parameter
that can be set by the user in the call to SLOPE()
.
Gaussian
This penalty sequence is related to BH, such that $$ \lambda_i = \lambda^{(\mathrm{BH})}_i \sqrt{1 + w(i-1)\cdot \mathrm{cumsum}(\lambda^2)_i}, $$ for \(i=1,\dots,p\), where \(w(k) = 1/(n-k-1)\). We let \(\lambda_1 = \lambda^{(\mathrm{BH})}_1\) and adjust the sequence to make sure that it's non-increasing. Note that if \(p\) is large relative to \(n\), this option will result in a constant sequence, which is usually not what you would want.
OSCAR
This sequence comes from Bondell and Reich and is a linearly non-increasing sequence such that $$ \lambda_i = q(p - i) + 1. $$ for \(i = 1,\dots,p\).
There are currently two solvers available for SLOPE: FISTA (Beck and
Teboulle 2009) and ADMM (Boyd et al. 2008). FISTA is available for
families but ADMM is currently only available for family = "gaussian"
.
SLOPE()
solves the convex minimization problem
$$
f(\beta) + \alpha \sum_{i=j}^p \lambda_j |\beta|_{(j)},
$$
where \(f(\beta)\) is a smooth and convex function and
the second part is the sorted L1-norm.
In ordinary least-squares regression,
\(f(\beta)\) is simply the squared norm of the least-squares residuals.
See section Families for specifics regarding the various types of
\(f(\beta)\) (model families) that are allowed in SLOPE()
.
By default, SLOPE()
fits a path of models, each corresponding to
a separate regularization sequence, starting from
the null (intercept-only) model to an almost completely unregularized
model. These regularization sequences are parameterized using
\(\lambda\) and \(\alpha\), with only \(\alpha\) varying along the
path. The length of the path can be manually, but will terminate
prematurely depending on
arguments tol_dev_change
, tol_dev_ratio
, and max_variables
.
This means that unless these arguments are modified, the path is not
guaranteed to be of length path_length
.
Bogdan, M., van den Berg, E., Sabatti, C., Su, W., & Cand<U+00E8>s, E. J. (2015). SLOPE -- adaptive variable selection via convex optimization. The Annals of Applied Statistics, 9(3), 1103<U+2013>1140. https://doi.org/10/gfgwzt
Bondell, H. D., & Reich, B. J. (2008). Simultaneous Regression Shrinkage, Variable Selection, and Supervised Clustering of Predictors with OSCAR. Biometrics, 64(1), 115<U+2013>123. JSTOR. https://doi.org/10.1111/j.1541-0420.2007.00843.x
Boyd, S., Parikh, N., Chu, E., Peleato, B., & Eckstein, J. (2010). Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers. Foundations and Trends<U+00AE> in Machine Learning, 3(1), 1<U+2013>122. https://doi.org/10.1561/2200000016
Beck, A., & Teboulle, M. (2009). A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems. SIAM Journal on Imaging Sciences, 2(1), 183<U+2013>202. https://doi.org/10.1137/080716542
plot.SLOPE()
, plotDiagnostics()
, score()
, predict.SLOPE()
,
trainSLOPE()
, coef.SLOPE()
, print.SLOPE()
, print.SLOPE()
,
deviance.SLOPE()
# NOT RUN {
# Gaussian response, default lambda sequence
fit <- SLOPE(bodyfat$x, bodyfat$y)
# Poisson response, OSCAR-type lambda sequence
fit <- SLOPE(abalone$x,
abalone$y,
family = "poisson",
lambda = "oscar",
q = 0.4)
# Multinomial response, custom alpha and lambda
m <- length(unique(wine$y)) - 1
p <- ncol(wine$x)
alpha <- 0.005
lambda <- exp(seq(log(2), log(1.8), length.out = p*m))
fit <- SLOPE(wine$x,
wine$y,
family = "multinomial",
lambda = lambda,
alpha = alpha)
# }
Run the code above in your browser using DataCamp Workspace