Learn R Programming

latrend (version 1.6.1)

A Framework for Clustering Longitudinal Data

Description

A framework for clustering longitudinal datasets in a standardized way. The package provides an interface to existing R packages for clustering longitudinal univariate trajectories, facilitating reproducible and transparent analyses. Additionally, standard tools are provided to support cluster analyses, including repeated estimation, model validation, and model assessment. The interface enables users to compare results between methods, and to implement and evaluate new methods with ease. The 'akmedoids' package is available from .

Copy Link

Version

Install

install.packages('latrend')

Monthly Downloads

405

Version

1.6.1

License

GPL (>= 2)

Issues

Pull Requests

Stars

Forks

Repository

https://github.com/philips-software/latrend

Homepage

https://philips-software.github.io/latrend/

Maintainer

Niek Den Teuling

Last Published

May 15th, 2024

Functions in latrend (1.6.1)

as.data.frame.lcModels

Generate a data.frame containing the argument values per method per row

as.list.lcMethod

Extract the method arguments as a list

Average posterior probability of assignment (APPA)

Odds of correct classification (OCC)

Biweekly Mean PAP Therapy Adherence of OSA Patients over 1 Year

Number of trajectories per cluster

lcMethod estimation step: compose an lcMethod object

as.data.frame.lcMethod

Convert lcMethod arguments to a list of atomic types

Convert a list of lcMethod objects to a lcMethods list

confusionMatrix

Compute the posterior confusion matrix

Update the cluster names

clusterTrajectories

Extract cluster trajectories

clusterProportions

Proportional size of each cluster

Check model convergence

createTrainDataFolds

Create the training data for each of the k models in k-fold cross validation evaluation

.defineInternalDistanceMetrics

Define the distance metrics for multiple types at once

Weekly Mean PAP Therapy Usage of OSA Patients in the First 3 Months

latrend-specific assertions

defineInternalMetric

Define an internal metric for lcModels

Get the cluster names

Extract lcModel coefficients

Select trajectories

createTestDataFold

Create the test fold data for validation

df.residual.lcModel

Extract the residual degrees of freedom from a lcModel

createTestDataFolds

Create all k test folds from the training data

.guessResponseVariable

Guess the response variable

evaluate.lcMethod

Substitute the call arguments for their evaluated values

deviance.lcModel

lcModel deviance

Estimation time

formula.lcModel

Extract the formula of a lcModel

formula.lcMethod

Extract formula

generateLongData

Generate longitudinal test data

getArgumentDefaults

Default argument values for the given method specification

fittedTrajectories

Extract the fitted trajectories

Extract lcModel fitted values

Get the method specification

defineExternalMetric

Define an external metric for lcModels

lcMethod estimation step: logic for fitting the method to the processed data

getExternalMetricNames

Get the names of the available external metrics

getInternalMetricDefinition

Get the internal metric definition

Compute external model metric(s)

Get citation info

interface-crimCV

crimCV interface

getInternalMetricNames

Get the names of the available internal metrics

Extract the trajectory identifier variable

interface-akmedoids

akmedoids interface

Get the trajectory ids on which the model was fitted

[[,lcMethod-method

Retrieve and evaluate a lcMethod argument by name

initialize,lcMethod-method

lcMethod initialization

interface-flexmix

flexmix interface

getArgumentExclusions

Arguments to be excluded from the specification

getExternalMetricDefinition

Get the external metric definition

interface-featureBased

featureBased interface

interface-mixtools

mixtools interface

getCall.lcModel

Get the model call

Check if object is of Class

interface-mixtvem

mixtvem interface

interface-funFEM

funFEM interface

clusterTrajectories,lcModelPartition-method

function interface

getArgumentDefaults,lcMethodLcmmGMM-method

interface-dtwclust

dtwclust interface

Check whether the argument of a lcMethod has a defined value.

latrend-approaches

High-level approaches to longitudinal clustering

latrend-methods

Supported methods for longitudinal clustering

latrend-metrics

Longitudinal dataset representation

interface-mixAK

mixAK interface

interface-metaMethods

lcMetaMethod abstract class

Cluster longitudinal data using the specified method

Cluster longitudinal data for a list of method specifications

interface-mclust

mclust interface

Cluster longitudinal data using bootstrapping

latrend-package

latrend: A Framework for Clustering Longitudinal Data

latrend-generics

Generics used by latrend for different classes

Method fit modifiers

latrend-estimation

Overview of lcMethod estimation functions

lcApproxModel-class

lcApproxModel class

lcMethodFunction

Specify a custom method based on a function

lcMethod-estimation

Longitudinal cluster method (lcMethod) estimation procedure

lcMethodAkmedoids

Specify AKMedoids method

Specify a zero-inflated repeated-measures GBTM method

lcMethodFeature

Feature-based clustering

lcMethodFlexmix

Method interface to flexmix()

Cluster longitudinal data over k folds

lcMethodDtwclust

Specify time series clustering via dtwclust

lcMatrixMethod-class

Cluster longitudinal data repeatedly

Artificial longitudinal dataset comprising three classes

latrend-parallel

Parallel computation using latrend

lcMethodFlexmixGBTM

Group-based trajectory modeling using flexmix

lcMethodMclustLLPA

Longitudinal latent profile analysis

lcMethodMixAK_GLMM

Specify a GLMM iwht a normal mixture in the random effects

Specify a FunFEM method

lcMethodLcmmGBTM

Specify GBTM method

lcMethodLcmmGMM

Specify GMM method using lcmm

lcMethodMixtoolsGMM

Specify mixed mixture regression model using mixtools

Two-step clustering through latent growth curve modeling and k-means

lcMethodMixTVEM

Specify a MixTVEM

lcModel class

lcModels: a list of lcModel objects

Construct a list of lcModel objects

lcModel-data-filters

Data filters for lcModel

Specify a longitudinal k-means (KML) method

lcModelPartition

Create a lcModel with pre-defined partitioning

lcModelWeightedPartition

Create a lcModel with pre-defined weighted partitioning

Select the lcModel with the lowest metric value

Specify a random-partitioning method

Select the lcModel with the highest metric value

Number of clusters

Cluster-handling functions for lcModel implementations.

lcMethodMixtoolsNPRM

Specify non-parametric estimation for independent repeated measures

Longitudinal cluster result (lcModel)

Compute internal model metric(s)

Two-step clustering through linear regression modeling and k-means

Number of trajectories

Extract the model training data

model.frame.lcModel

Extract model training data

names,lcMethod-method

lcMethod argument names

model.data.lcModel

Extract the model data that was used for fitting

plotTrajectories

Plot the data trajectories

lcMethodStratify

Specify a stratification method

plotFittedTrajectories

Plot the fitted trajectories

Plot one or more internal metrics for all lcModels

Generate a list of lcMethod objects

Extract the log-likelihood of a lcModel

Argument matching with defaults and parent ellipsis expansion

Number of observations used for the lcModel fit

plot-lcModel-method

Mean ignoring NAs

plotClusterTrajectories

Plot cluster trajectories

plot-lcModels-method

Grid plot for a list of models

Posterior probability per fitted trajectory

postProbFromObs

Compute the id-specific postprob matrix from a given observation-level postprob matrix

postprobFromAssignments

Create a posterior probability matrix from a vector of cluster assignments.

lcMethod estimation step: method preparation logic

predictAssignments

Predict the cluster assignments for new trajectories

predict.lcModel

lcModel predictions

lcMethod estimation step: logic for post-processing the fitted lcModel

residuals.lcModel

Extract lcModel residuals

predictPostprob

Posterior probability for new data

predictForCluster

Predict trajectories conditional on cluster membership

Print the arguments of an lcMethod object

Print lcModels list concisely

Quantile-quantile plot

lcMethod estimation step: logic for preparing the training data

Test a condition

subset.lcModels

Subsetting a lcModels list based on method arguments

summary.lcModel

Summarize a lcModel

Test the implementation of an lcMethod and associated lcModel subclasses

Sampling times of a lcModel

responseVariable

Extract response variable

Extract the time variable

Reduce the memory footprint of an object for serialization

Extract residual standard deviation from a lcModel

Convert a multiple time series matrix to a data.frame

transformFitted

Helper function for custom lcModel classes implementing fitted.lcModel()

lcMethod estimation step: method argument validation logic

trajectoryAssignments

Get the cluster membership of each trajectory

Convert a longitudinal data.frame to a matrix

Update a lcModel

update.lcMethod

Update a method specification

Get the trajectories

weighted.meanNA

Weighted arithmetic mean ignoring NAs

transformPredict

Helper function for custom lcModel classes implementing predict.lcModel()

Sample an index of a vector weighted by the elements

as.data.frame.lcMethods

Convert a list of lcMethod objects to a data.frame

Convert a list of lcModels to a lcModels list