Q_theta

nobs

a user supplied lambda sequence. Typically, by leaving this
option unspecified users can have the program compute its own lambda
sequence based on <code>nlambda</code> and <code>lambda.factor</code>. Supplying a
value of lambda overrides this. It is better to supply a decreasing
sequence of lambda values than a single (small) value, if not, the program
will sort user-defined lambda sequence in decreasing order automatically.
Default: <code>NULL</code>.

lambda

the mixing tuning parameter, with $0&lt;\alpha&lt;1$. It controls
the penalization strength between the main effects and the interactions.
The penalty is defined as $$\lambda(1-\alpha)(w_e|\beta_e|+ \sum w_j
||\beta_j||_2) + \lambda\alpha(\sum w_{je} |\gamma_j|)$$Larger values of
<code>alpha</code> will favor selection of main effects over interactions.
Smaller values of <code>alpha</code> will allow more interactions to enter the
final model. Default: <code>0.5</code>

alpha

penalty factor for exposure variable

betaE

theta_list

gamma

calculates likelihood function. Used to assess convergence of
 fitting algorithm. This corresponds to the Q(theta) function in the paper

Sparse additive interaction learning with the strong heredity property, i.e.,
an interaction is selected only if its corresponding main effects are also included.
Fits a linear model with non-linear interactions via penalized maximum likelihood.
Interactions are limited to a single exposure or environment variable. For more information,
see the website below and the accompanying paper: Bhatnagar et al., "A sparse additive model for
high-dimensional interactions with an exposure variable", 2019, <DOI:10.1101/445304>.

Q_theta: Objective function

Description

Usage

Arguments

Value