Time Series Representations
Creates bit-level (clipped representation) from a vector
Functions for linear regression model coefficients extraction
Arctangent denormalisation
Two-parameter Box-Cox denormalisation
Min-Max denormalisation
Yeo-Johnson denormalisation
Z-score denormalisation
Fast statistic functions (helpers)
MAAPE
MAE
MAPE
MASE
MdAE
MSE
Arctangent normalisation
Two-parameter Box-Cox normalisation
Min-Max normalisation
Min-Max normalization list
Min-Max normalisation with parameters
Yeo-Johnson normalisation
Z-score normalisation
Z-score normalization list
Z-score normalisation with parameters
DCT representation
DFT representation by FFT
DWT representation
Exponential smoothing seasonal coefficients as representation
FeaClip representation of time series
FeaClipTrend representation of time series
FeaTrend representation of time series
GAM regression coefficients as representation
Computation of list of representations list of time series with differ...
Regression coefficients from linear model as representation
Computation of matrix of representations from matrix of time series
PAA - Piecewise Aggregate Approximation
PIP representation
PLA representation
SAX - Symbolic Aggregate Approximation
Mean seasonal profile of time series
Simple Moving Average representation
Windowing of time series
RLE (Run Length Encoding) written in C++
RMSE
sMAPE
Creates bit-level (trending) representation from a vector
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.
Useful links