In andreas50/utsOperators: Moving Averages and Other Rolling Operators for Unevenly Spaced Time Series

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/"
)
options(width=100)
library(utsOperators)

Introduction

This package provides rolling time series operators for unevenly spaced data, such as simple moving averages (SMAs), exponential moving averages (EMAs), and arbitrary rolling R functions. It is a wrapper around the highly-optimized C library utsAlgorithms. The time series class used by this package is the uts class.

The package rcpputs is a low-level wrapper around the same C library that does not rely on any time series class, but instead requires the user to pass in a vector of observation values and observation times to each function.

Installation

This package is not yet available on CRAN, but can be installled from GitHub:

devtools::install_github(c("andreas50/uts", "andreas50/utsOperators"))  # using package 'devtools'
remotes::install_github(c("andreas50/uts", "andreas50/utsOperators"))   # ... or using package 'remotes'

Sample Code

# Get sample unevenly-spaced time series with six observations
x <- ex_uts()
x

# SMA with last-point interpolation, 1-day wide rolling time window
sma(x, ddays(1))

# EMA with linear interpolation, 12-hour effective temporal length
ema(x, dhours(12), interpolation="linear")

# Rolling mean, sum, number of observation values in a 1-day wide rolling time window
rolling_apply(ex_uts(), width=ddays(1), FUN=mean)
rolling_apply(ex_uts(), width=ddays(1), FUN=sum)
rolling_apply(ex_uts(), width=ddays(1), FUN=length)

andreas50/utsOperators documentation built on May 25, 2019, 7:16 a.m.