Learn R Programming

⚠️There's a newer version (1.6.1) of this package.Take me there.

latrend (version 1.1.0)

A Framework for Clustering Longitudinal Data

Description

A framework for clustering longitudinal datasets in a standardized way. 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.

Copy Link

Version

Install

install.packages('latrend')

Monthly Downloads

421

Version

1.1.0

License

GPL (>= 2)

Issues

Pull Requests

Stars

Forks

Repository

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

Maintainer

Niek Den Teuling

Last Published

March 5th, 2021

Functions in latrend (1.1.0)

as.data.frame.lcModels

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

Convert a list of lcMethod objects to a lcMethods list

Update the cluster names

Get the cluster names

interface-custom

custom interface

clusterTrajectories

Extract the cluster trajectories

externalMetric,lcModel,lcModel-method

Compute external model metric(s)

confusionMatrix

Compute the posterior confusion matrix

Check model convergence

getInternalMetricDefinition

Get the internal metric definition

getInternalMetricNames

Get the names of the available internal metrics

interface-funFEM

funFEM interface

Extract lcModel fitted values

Coefficients of a lcModel

lcApproxModel-class

lcApproxModel class

interface-mixAK

mixAK interface

interface-mclust

mclust interface

interface-dtwclust

dtwclust interface

latrend-specific assertions

formula.lcMethod

Extract formula

Specify a longitudinal k-means (KML) method

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

lcMatrixMethod-class

lcMethodMixtoolsGMM

Specify mixed mixture regression model using mixtools

formula.lcModel

Extract the formula of a lcModel

lcMethodDtwclust

Specify time series clustering via dtwclust

Create a lcMethod object of the specified type and arguments

lcMethodFeature

Feature-based clustering

Specify a random-partitioning method

getExternalMetricNames

Get the names of the available external metrics

getExternalMetricDefinition

Get the external metric definition

interface-crimCV

crimCV interface

$,lcMethod-method

Retrieve and evaluate a lcMethod argument by name

lcMethodStratify

Specify a stratification method

Specify a model based on a pre-computed result.

lcModelPartition

Create a lcModel with pre-defined partitioning

interface-akmedoids

akmedoids interface

model.frame.lcModel

Extract model training data

interface-mixtools

mixtools interface

interface-mixtvem

mixtvem interface

Convert a list of lcModels to a lcModels list

Extract the log-likelihood of a lcModel

lcMethodMixtoolsNPRM

Specify non-parametric estimation for independent repeated measures

Argument matching with defaults and parent ellipsis expansion

Cluster-handling functions for lcModel implementations.

lcModel-data-filters

Data filters for lcModel

Extract the model training data

as.data.frame.lcMethods

Convert a list of lcMethod objects to a data.frame

as.data.frame.lcMethod

Convert lcMethod arguments to a list of atomic types

model.data.lcModel

Extract the model data that was used for fitting

postprobFromAssignments

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

Number of clusters

plot,lcModel,ANY-method

plotClusterTrajectories

Plot cluster trajectories

as.list.lcMethod

Extract the method arguments as a list

Number of strata

predict.lcModel

lcModel predictions

Extract the number of observations from a lcModel

Extract residual standard deviation from a lcModel

Strip a lcModel for serialization

residuals.lcModel

Extract lcModel residuals

createTestDataFolds

Create all k test folds from the training data

createTestDataFold

Create the test fold data for validation

defineExternalMetric

Define an external metric for lcModels

defineInternalMetric

Define an internal metric for lcModels

generateLongData

Generate longitudinal test data

getCall.lcModel

Get the model call

subset.lcModels

Subsetting a lcModels list based on method arguments

responseVariable

Extract the response variable

interface-featureBased

featureBased interface

summary.lcModel

Summarize a lcModel

dcastRepeatedMeasures

Cast a longitudinal data.frame to a matrix

Get the method specification of a lcModel

createTrainDataFolds

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

Get the model estimation time

evaluate.lcMethod

Substitute the call arguments for their evaluated values

transformFitted

Helper function for ensuring the right fitted() output

clusterProportions

Proportional size of each cluster

Number of strata per cluster

Extract the trajectory identifier variable

Cluster longitudinal data

transformLatrendData

Transform latrend input data into the right format

Get the unique ids included in this model

df.residual.lcModel

Extract the residual degrees of freedom from a lcModel

interface-longclust

longclust interface

deviance.lcModel

lcModel deviance

[[,lcMethod-method

Retrieve and evaluate a lcMethod argument by name

getName,lcMethodLcmmGMM-method

Cluster longitudinal data for a list of model specifications

Create a lcMethod object from a call

Check if object is of Class

lcMethodFlexmix

Method interface to flexmix()

lcMethodLcmmGBTM

Specify GBTM method

lcMethodAkmedoids

Specify AKMedoids method

lcMethodFlexmixGBTM

Group-based trajectory modeling using flexmix

lcMethodLcmmGMM

Specify GMM method using lcmm

Compute internal model metric(s)

interface-flexmix

flexmix interface

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

Synthetic longitudinal dataset comprising three classes

Specify a zero-inflated repeated-measures GBTM method

Cluster longitudinal data repeatedly

latrend-package

latrend: A Framework for Clustering Longitudinal Data

Posterior probability per fitted id

latrend-generics

Method- and model-specific generics defined by the latrend package

Select the lcModel with the lowest metric value

Print lcModels list concisely

Specify a custom method based on a model function

postProbFromObs

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

transformPredict

Helper function that matches the output to the specified newdata

Generate a list of lcMethod objects

lcMethodLongclust

Specify Longclust method

lcMethodMclustLLPA

Longitudinal latent profile analysis

Quantile-quantile plot

Update a lcModel

weighted.meanNA

Weighted arithmetic mean ignoring NAs

Sample an index of a vector weighted by the elements

Cluster longitudinal data using bootstrapping

Extract the fitted trajectories for all strata

predictAssignments

Predict the cluster assignments for new trajectories

predictForCluster

lcModel prediction for a specific cluster

Cluster longitudinal data over k folds

Specify a FunFEM method

update.lcMethod

Update a method specification

trajectoryAssignments

Get the cluster membership of each trajectory

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

lcMethodMixAK_GLMM

Specify a GLMM iwht a normal mixture in the random effects

lcModelWeightedPartition

Create a lcModel with pre-defined weighted partitioning

lcMethodMixTVEM

Specify a MixTVEM

Construct a flat (named) list of lcModel objects

Select the lcModel with the highest metric value

meltRepeatedMeasures

Convert a repeated measures data matrix to a data.frame

Plot one or more internal metrics for all lcModels

predictPostprob

lcModel posterior probability prediction

plotTrajectories

Plot trajectories

Print the arguments of an lcMethod object

Sampling times of a lcModel

Extract the time variable