dlm: Distributed lag models

In Biostatistics4SocialImpact/dlm: Distributed Lag Models for Built Environment Applications

Description

Fit distributed lag models using lme4 to penalize smooth terms. Other random effects terms and generalized mixed models supported.

Usage

dlm(formula, data, subset, na.action, weights, offset,
    method = c("REML", "MLE"), family = gaussian(),
    control = list(), ...)

Arguments

formula	an object of class stats::formula: a symbolic description of the model to be fitted. See Details
data	an optional data frame, list, or environment containing the variables of the model to be fitted
subset	optional vector specifying a subset of observations to be used in the fitting process
na.action	optional function that indicates what should happen when the data contains NA's. The default is set by the na.action setting of base::options, usually stats::na.omit
weights	optional vector of weights to be used in the fitting process. Should be NULL or a numeric vector
offset	a known offset term to include in the model, e.g. for poisson() family models
method	algorithm used to fit the DLM. Partial matching and capitalization allowed. The default is "REML" for linear/gaussian(link = "identity") family models, and "MLE" otherwise
family	a description of the error distribution and link function to be used in the model. The default is gaussian(link = "identity"). See stats::family for possible family functions and details
control	either a list object with arguments to be passed to the lme4::lmerControl sequence, or the output of [g]lmerConrol directly
...	Additional parameters passed to lme4::lFormula

Details

Models are specified using typical lme4 formula syntax with at least one set of lag terms returned by a given smoothing function (e.g. see cr). The smoothing function can be any that returns a SmoothLag basis object. See Examples for a basic call to dlm using the formula interface, and a cubic radial lag basis specified via cr, and the dlmBE package documentation for a discussion of the types of models dlm is designed to handle.

Value

An S4 object that inherits from dlMod (and lme4::merMod, by extension) containing the results of the fitted model. Many standard model summary methods are available for these object types

References

Baek J, Sanchez BN, Berrocal VJ, & Sanchez-Vaznaugh EV (2016) Epidemiology 27(1):116-24. (PubMed)

Baek J, Hirsch JA, Moore K, Tabb LP, et al. (2017) Epidemiology 28(3):403-11. (PubMed)

Bates D, Maechler M, Bolker BM, & Walker SC (2015) Fitting linear mixed-effects models using lme4. J Stat Softw 67(1). (jstatsoft.org)

See Also

lme4::lmer, cr, dlMod

Examples

data (simdata)

## Setup distance count matrix and corresponding lag distances
X <- as.matrix(simdata[, -(1:3)])
lag <- seq(0.1, 10, length.out = ncol(X))

fit <- dlm(Y ~ Age + Gender + cr(lag, X), data = simdata)
summary (fit)

Biostatistics4SocialImpact/dlm documentation built on May 19, 2019, 10:47 p.m.