ts_fs_taxonomy_results: A data frame of 1000 different classified time series data.

In mowomoyela/tsfcmethodr: Classification methods to predict best performing forecasting method of a ts

Description

A data frame containing 1000 different classified univariate and multivariate time series data from the tstaxonomyr R package. Each row represents one single time series that is classified on XX different (statistical) features which are listed in the following item description. This time the time series data was classified on the feature selection reduced taxonomy. All features/columns are factor values of the different classification classes.

Usage

ts_fs_taxonomy_results

Format

A data frame with 1000 elements. Each row represents a time series object that is classified on the following 15 (statistical features) from the tstaxonomyr R package representing the columns of the data frame. This time the time series data was classified on the feature selection reduced taxonomy:

Trend

The trend factor of the time series

Autocorrelation

The autocorrelation factor of the time series

Seasonality

The seasonality factor of the time series

Periodicity

The periodicity factor of the time series

Chaos

The chaos factor of the time series

Entropy

The entropy factor of the time series

TurningPoints

The number of turning points of the time series

PartialAutocorrelation

The partial autocorrelation factor of the time series

Outliers

The number of outliers of the time series

StepChanges

The step changes factor of the time series

Peaks

The number of peaks of the time series

DurbinWatsonTest

The dw test factor of the time series

calculate_determination_coefficient

The R^2 factor of the time series

NumberOfAttributes

The number of features of the time series

ts_name

The name of the time series from the tstaxonomyr package

Source

tstaxonomyr R package https://github.com/mowomoyela/tstaxonomyr

mowomoyela/tsfcmethodr documentation built on May 15, 2019, 4:47 p.m.