TSrepr1.1.0 package

Time Series Representations

clipping

Creates bit-level (clipped representation) from a vector

coef_comp

Functions for linear regression model coefficients extraction

denorm_atan

Arctangent denormalisation

denorm_boxcox

Two-parameter Box-Cox denormalisation

denorm_min_max

Min-Max denormalisation

denorm_yj

Yeo-Johnson denormalisation

denorm_z

Z-score denormalisation

fast_stat

Fast statistic functions (helpers)

maape

MAAPE

mae

MAE

mape

MAPE

mase

MASE

mdae

MdAE

mse

MSE

norm_atan

Arctangent normalisation

norm_boxcox

Two-parameter Box-Cox normalisation

norm_min_max

Min-Max normalisation

norm_min_max_list

Min-Max normalization list

norm_min_max_params

Min-Max normalisation with parameters

norm_yj

Yeo-Johnson normalisation

norm_z

Z-score normalisation

norm_z_list

Z-score normalization list

norm_z_params

Z-score normalisation with parameters

repr_dct

DCT representation

repr_dft

DFT representation by FFT

repr_dwt

DWT representation

repr_exp

Exponential smoothing seasonal coefficients as representation

repr_feaclip

FeaClip representation of time series

repr_feacliptrend

FeaClipTrend representation of time series

repr_featrend

FeaTrend representation of time series

repr_gam

GAM regression coefficients as representation

repr_list

Computation of list of representations list of time series with differ...

repr_lm

Regression coefficients from linear model as representation

repr_matrix

Computation of matrix of representations from matrix of time series

repr_paa

PAA - Piecewise Aggregate Approximation

repr_pip

PIP representation

repr_pla

PLA representation

repr_sax

SAX - Symbolic Aggregate Approximation

repr_seas_profile

Mean seasonal profile of time series

repr_sma

Simple Moving Average representation

repr_windowing

Windowing of time series

rleC

RLE (Run Length Encoding) written in C++

rmse

RMSE

smape

sMAPE

trending

Creates bit-level (trending) representation from a vector

TSrepr

TSrepr package

Methods for representations (i.e. dimensionality reduction, preprocessing, feature extraction) of time series to help more accurate and effective time series data mining. Non-data adaptive, data adaptive, model-based and data dictated (clipped) representation methods are implemented. Also various normalisation methods (min-max, z-score, Box-Cox, Yeo-Johnson), and forecasting accuracy measures are implemented.

  • Maintainer: Peter Laurinec
  • License: GPL-3 | file LICENSE
  • Last published: 2020-07-13