dlnm-internal: Internal Functions for Package dlnm

In gasparrini/dlnm: Distributed Lag Non-Linear Models

Description

These functions are called internally by other functions and are not meant to be directly run by the users.

Usage

fci(ci, x, high, low, ci.arg, plot.arg, noeff=NULL)

findrank(X)

getcoef(model, class)

getlink(model, class, model.link=NULL)

getvcov(model, class)

mkaddSlag(addSlag, d)

mkat(at, from, to, by, range, lag, bylag)

mkcen(cen, type, basis, range)

mklag(lag)

mkXpred(type, basis, at, predvar, predlag, cen)

seqlag(lag, by=1)

Arguments

ci	type of confidence intervals representation.
x	the coordinates of the x axis.
high, low	the coordinates of the y axis for the interval.
ci.arg	list of arguments to draw the confidence intervals.
plot.arg	list of arguments of the main plot.
noeff	reference value of the null effect.
X	a matrix.
model	a regression model object.
class	a regression model class.
addSlag	matrix or vector (or list of matrices and/or vectors) defining additional penalties on the lag structure.
d	numeric vector of length 2 providing the cross-basis dimensions.
at	either a numeric vector representing the values of a constant exposure throughout the lag period defined by lag, or a matrix of exposure histories over the same lag period used for estimation.
from, to	range of predictor values used for prediction.
by, bylag	increment of the sequences of predictor and lag values used for prediction.
range	range of values used for prediction.
lag	either an integer scalar or vector of length 2, defining the the maximum lag or the lag range, respectively.
cen	logical or a numeric scalar. It specifies the centering value, then used as a reference for predictions.
type, basis	type of model and related basis object from which predictions are needed. See crosspred.
predvar	vector or matrix of predictor values used for prediction
predlag	vector or matrix of lag values used for prediction

Details

The function fci provides different options for representing confidence intervals, and it is called internally in plotting functions.

The function findrank returns the rank of a matrix.

The functions getcoef, getlink, and getvcov extract coefficients, the model link, and (co)variance matrix, respectively, depending on the class of the model, and returns a message error if the process fails. They are used internally in crosspred and crossreduce.

The function mkaddSlag returns a list of rescaled penalty matrices defining additional penalties on the lag structure. It is used intenally in functions for performing penalized models.

The function mkat is used internally in crosspred to define the values used for predictions.

The function mkcen is used internally in crosspred to define the centering value for computing predictions.

The functions mklag is used internally in several other functions to check/define the vector of length 2 representing the lag interval.

The function mkXpred is used internally in crosspred to define the basis or cross-basis matrix for computing predictions.

The function seqlag is used internally in several other functions to create the vector with the sequence of lags given the range provided.

Note

These and other undocumented internal functions are not exported in the namespace of the package dlnm. The user can visualize them through the use of the triple colon operator ':::' or through the function getAnywhere (see the examples below).

Author(s)

Antonio Gasparrini <[e-mail]>

See Also

See dlnm-package for an introduction to the package and for links to package vignettes providing more detailed information.

Examples

### display internal functions
dlnm:::getcoef
getAnywhere(getcoef)

### display other undocumented functions
dlnm:::fci
getAnywhere(fci)

gasparrini/dlnm documentation built on Oct. 9, 2021, 2:22 a.m.