RemixAutoML (version 0.5.4)

AutoDiffLagN: AutoDiffLagN

Description

AutoDiffLagN create differences for selected numerical columns

Usage

AutoDiffLagN(
  data,
  DateVariable = NULL,
  GroupVariables = NULL,
  DiffVariables = NULL,
  DiffDateVariables = NULL,
  DiffGroupVariables = NULL,
  NLag1 = 0L,
  NLag2 = 1L,
  Sort = FALSE,
  RemoveNA = TRUE
)

Arguments

data

Source data

DateVariable

Date column used for sorting

GroupVariables

Difference data by group

DiffVariables

Column names of numeric columns to difference

DiffDateVariables

Columns names for date variables to difference. Output is a numeric value representing the difference in days.

DiffGroupVariables

Column names for categorical variables to difference. If no change then the output is 'No_Change' else 'New=NEWVAL Old=OLDVAL' where NEWVAL and OLDVAL are placeholders for the actual values

NLag1

If the diff calc, we have column 1 - column 2. NLag1 is in reference to column 1. If you want to take the current value minus the previous weeks value, supply a zero. If you want to create a lag2 - lag4 NLag1 gets a 2.

NLag2

If the diff calc, we have column 1 - column 2. NLag2 is in reference to column 2. If you want to take the current value minus the previous weeks value, supply a 1. If you want to create a lag2 - lag4 NLag1 gets a 4.

Sort

TRUE to sort your data inside the function

RemoveNA

Set to TRUE to remove rows with NA generated by the lag operation

See Also

Other Feature Engineering: AutoDataPartition(), AutoHierarchicalFourier(), AutoInteraction(), AutoLagRollStatsScoring(), AutoLagRollStats(), AutoTransformationCreate(), AutoTransformationScore(), AutoWord2VecModeler(), AutoWord2VecScoring(), CategoricalEncoding(), CreateCalendarVariables(), CreateHolidayVariables(), DummifyDT(), H2OAutoencoderScoring(), H2OAutoencoder(), ModelDataPrep(), TimeSeriesFill()

Examples

# NOT RUN {
# Create fake data
data <- RemixAutoML::FakeDataGenerator(
  Correlation = 0.70,
  N = 50000,
  ID = 2L,
  FactorCount = 3L,
  AddDate = TRUE,
  ZIP = 0L,
  TimeSeries = FALSE,
  ChainLadderData = FALSE,
  Classification = FALSE,
  MultiClass = FALSE)

# Store Cols to diff
Cols <- names(data)[which(unlist(data[, lapply(.SD, is.numeric)]))]

# Clean data before running AutoDiffLagN
data <- RemixAutoML::ModelDataPrep(data = data, Impute = FALSE, CharToFactor = FALSE, FactorToChar = TRUE)

# Run function
data <- RemixAutoML::AutoDiffLagN(
  data,
  DateVariable = "DateTime",
  GroupVariables = c("Factor_1", "Factor_2"),
  DiffVariables = Cols,
  DiffDateVariables = NULL,
  DiffGroupVariables = NULL,
  NLag1 = 0L,
  NLag2 = 1L,
  Sort = TRUE,
  RemoveNA = TRUE)
# }
# NOT RUN {
# }