SMG.Lag: SMG.Lag

In frbl/OnlineSuperLearner: Online SuperLearner package

Description

The OnlineSuperLearner uses various SummaryMeasureGenerator instances to generate new variables that can be used to represent the relevant history of a variable. The SMG.Lag class allows a user to specify lagged variables. That is, in the prediction of a variable at time $t$, one can include variables from past measurements, if there is an lagged or cross lagged effect.

Usage

SMG.Lag

Format

An object of class R6ClassGenerator of length 24.

Methods

initialize(lags, colnames.to.lag)

Initializes a new SMG.Lag object.

@param lags integer the number of lags the SMG.Lag object should create (if there is a lag_1 variable, this number should be 1).

@param colnames.to.lag vector a vector containg the names of the variables to lag.

lagged_colnames(colnames = NULL)

This function returns a list of names that represent the lagged names of the colnames provided to it. It uses the lags variable as specified in the initialization step. One can choose to provide colnames to this function to get the result for a specific set of colnames. If no colnames are provided, the function uses the colnames.to.lag names as specified on initialization.

@param colnames vector (default = NULL) the vector of names for which to genearte a lagged version.

@return vector a vector of colnames with appended '_lag_x' names.

update(data.current)

For online learning we need to be able to create new data blocks on the fly (as not all data is available beforehand. This function updates a set of variables to contain the lagged variables for the next block.

@param data.current data.table the current data / the last data used for training, from which a new summary measure needs to be generated.

@return data.table a new block for the next training iteration.

process(data.current)

Converts lagged columns for all provided data. It goes through the data and generates the lagged variables accordingly.

@param data.current data.table the currently available data.

@return data.table with the new summary measure columns.

minimalObservations

Active method. The minimal number of measurements needed for this SMG to be able to generate a new block. This is the number of lags + 1.

@return integer the number of lags + 1

exposedVariables

Active method. Returns a list of variables returned by this SMG.

@return vector a vector of strings conaining all lagged columnames.

frbl/OnlineSuperLearner documentation built on Feb. 9, 2020, 9:28 p.m.