tsfeatures v1.0.1
0
Monthly downloads
Time Series Feature Extraction
Methods for extracting various features from time series data. The features provided are those from Hyndman, Wang and Laptev (2013) <doi:10.1109/ICDMW.2015.104>, Kang, Hyndman and Smith-Miles (2017) <doi:10.1016/j.ijforecast.2016.09.004> and from Fulcher, Little and Jones (2013) <doi:10.1098/rsif.2013.0048>. Features include spectral entropy, autocorrelations, measures of the strength of seasonality and trend, and so on. Users can also define their own feature functions.
Readme
tsfeatures
The R package tsfeatures provides methods for extracting various features from time series data.
Installation
You can install the stable version on R CRAN.
install.packages('tsfeatures', dependencies = TRUE)
You can install the development version from Github with:
# install.packages("devtools")
devtools::install_github("robjhyndman/tsfeatures")
Usage
library(tsfeatures)
mylist <- list(sunspot.year, WWWusage, AirPassengers, USAccDeaths)
myfeatures <- tsfeatures(mylist)
myfeatures
#> # A tibble: 4 x 20
#> frequency nperiods seasonal_period trend spike linearity curvature
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 0 1 0.125 2.10e-5 3.58 1.11
#> 2 1 0 1 0.985 3.01e-8 4.45 1.10
#> 3 12 1 12 0.989 2.12e-8 11.0 1.10
#> 4 12 1 12 0.796 9.67e-7 -2.13 2.85
#> # ... with 13 more variables: e_acf1 <dbl>, e_acf10 <dbl>, entropy <dbl>,
#> # x_acf1 <dbl>, x_acf10 <dbl>, diff1_acf1 <dbl>, diff1_acf10 <dbl>,
#> # diff2_acf1 <dbl>, diff2_acf10 <dbl>, seasonal_strength <dbl>,
#> # peak <dbl>, trough <dbl>, seas_acf1 <dbl>
License
This package is free and open source software, licensed under GPL-3.
Functions in tsfeatures
| Name | Description | |
| hurst | Hurst coefficient | |
| ac_9 | Autocorrelation at lag 9. Included for completion and consistency. | |
| compengine | CompEngine feature set | |
| binarize_mean | Converts an input vector into a binarized version from software package hctsa | |
| acf_features | Autocorrelation-based features | |
| fluctanal_prop_r1 | Implements fluctuation analysis from software package hctsa | |
| entropy | Spectral entropy of a time series | |
| max_level_shift | Time series features based on sliding windows | |
| unitroot_kpss | Unit Root Test Statistics | |
| firstmin_ac | Time of first minimum in the autocorrelation function from software package hctsa | |
| crossing_points | Number of crossing points | |
| dist_features | The distribution feature set from software package hctsa | |
| embed2_incircle | Points inside a given circular boundary in a 2-d embedding space from software package hctsa | |
| pred_features | The prediction feature set from software package hctsa | |
| lumpiness | Time series features based on tiled windows | |
| sampen_first | Second Sample Entropy of a time series from software package hctsa | |
| firstzero_ac | The first zero crossing of the autocorrelation function from software package hctsa | |
| yahoo_data | Yahoo server metrics | |
| walker_propcross | Simulates a hypothetical walker moving through the time domain from software package hctsa | |
| heterogeneity | Heterogeneity coefficients | |
| sampenc | Second Sample Entropy from software package hctsa | |
| flat_spots | Number of flat spots | |
| histogram_mode | Mode of a data vector from software package hctsa | |
| holt_parameters | Parameter estimates of Holt's linear trend method | |
| motiftwo_entro3 | Local motifs in a binary symbolization of the time series from software package hctsa | |
| nonlinearity | Nonlinearity coefficient | |
| outlierinclude_mdrmd | How median depend on distributional outliers from software package hctsa | |
| pacf_features | Partial autocorrelation-based features | |
| spreadrandomlocal_meantaul | Bootstrap-based stationarity measure from software package hctsa | |
| scal_features | The scaling feature set from software package hctsa | |
| station_features | The stationarity feature set from software package hctsa | |
| trev_num | Normalized nonlinear autocorrelation, the numerator of the trev function of a time series from software package hctsa | |
| arch_stat | ARCH LM Statistic | |
| as.list.mts | Convert mts object to list of time series | |
| tsfeatures | Time series feature matrix | |
| std1st_der | Standard deviation of the first derivative of the time series from software package hctsa | |
| stl_features | Strength of trend and seasonality of a time series | |
| autocorr_features | The autocorrelation feature set from software package hctsa | |
| localsimple_taures | The first zero crossing of the autocorrelation function of the residuals from Simple local time-series forecasting from software package hctsa | |
| No Results! | ||
Vignettes of tsfeatures
| Name | ||
| tsfeatures.Rmd | ||
| No Results! | ||
Last month downloads
Details
| License | GPL-3 |
| LazyData | true |
| ByteCompile | true |
| URL | https://pkg.robjhyndman.com/tsfeatures/ |
| BugReports | https://github.com/robjhyndman/tsfeatures/issues/ |
| RoxygenNote | 6.1.1 |
| VignetteBuilder | knitr |
| Encoding | UTF-8 |
| NeedsCompilation | no |
| Packaged | 2019-04-16 06:30:44 UTC; mitchell |
| Repository | CRAN |
| Date/Publication | 2019-04-16 13:02:47 UTC |
| suggests | dplyr , GGally , ggplot2 , knitr , Mcomp , rmarkdown , testthat , tidyr |
| imports | ForeCA , forecast (>= 8.3) , fracdiff , furrr , future , purrr , RcppRoll (>= 0.2.2) , stats , tibble , tseries , urca |
| depends | R (>= 3.2.3) |
| Contributors | Earo Wang, Nikolay Laptev, Yanfei Kang, Mitchell O'Hara-Wild, Thiyanga Talagala, Souhaib Ben Taieb, Yangzhuoran Yang, Pablo Montero-Manso, Cao Hanqing, D K Lake, J R Moorman |
Include our badge in your README
[](http://www.rdocumentation.org/packages/tsfeatures)