Repository Mirror for your Cloud Server and Webhosting

Type:

Package

Title:

Fast Computation of Running Statistics for Time Series

Version:

1.1.0

Description:

Provides methods for fast computation of running sample statistics for time series. These include: (1) mean, (2) standard deviation, and (3) variance over a fixed-length window of time-series, (4) correlation, (5) covariance, and (6) Euclidean distance (L2 norm) between short-time pattern and time-series. Implemented methods utilize Convolution Theorem to compute convolutions via Fast Fourier Transform (FFT).

License:

GPL-3

Encoding:

UTF-8

LazyData:

true

RoxygenNote:

6.1.1

URL:

https://github.com/martakarass/runstats

BugReports:

https://github.com/martakarass/runstats/issues

Imports:

fftwtools

Suggests:

covr, testthat, ggplot2, knitr, rmarkdown, sessioninfo, rbenchmark, cowplot, spelling

VignetteBuilder:

knitr

Language:

en-US

NeedsCompilation:

Packaged:

2019-11-14 19:59:37 UTC; martakaras

Author:

Marta Karas

[aut, cre], Jacek Urbanek

[aut], John Muschelli

[ctb], Lacey Etzkorn [ctb]

Maintainer:

Marta Karas <marta.karass@gmail.com>

Repository:

CRAN

Date/Publication:

2019-11-14 20:30:02 UTC

Fast Running Correlation Computation

Description

Computes running correlation between time-series x and short-time pattern y.

Usage

RunningCor(x, y, circular = FALSE)

Arguments

x

A numeric vector.

y

A numeric vector, of equal or shorter length than x.

circular

logical; whether running correlation is computed assuming circular nature of x time-series (see Details).

Details

Computes running correlation between time-series x and short-time pattern y. The length of output vector equals the length of x. Parameter circular determines whether x time-series is assumed to have a circular nature. Assume l_x is the length of time-series x, l_y is the length of short-time pattern y.

If circular equals TRUE then

first element of the output vector corresponds to sample correlation between x[1:l_y] and y,
last element of the output vector corresponds to sample correlation between c(x[l_x], x[1:(l_y - 1)]) and y.

If circular equals FALSE then

first element of the output vector corresponds to sample correlation between x[1:l_y] and y,
the l_x - W + 1-th element of the output vector corresponds to sample correlation between x[(l_x - l_y + 1):l_x],
last W-1 elements of the output vector are filled with NA.

See runstats.demo(func.name = "RunningCor") for a detailed presentation.

Value

A numeric vector.

Examples

x <- sin(seq(0, 1, length.out = 1000) * 2 * pi * 6)
y <- x[1:100]
out1 <- RunningCor(x, y, circular = TRUE)
out2 <- RunningCor(x, y, circular = FALSE)
plot(out1, type = "l"); points(out2, col = "red")

Fast Running Covariance Computation

Description

Computes running covariance between time-series x and short-time pattern y.

Usage

RunningCov(x, y, circular = FALSE)

Arguments

x

A numeric vector.

y

A numeric vector, of equal or shorter length than x.

circular

Logical; whether running variance is computed assuming circular nature of x time-series (see Details).

Details

Computes running covariance between time-series x and short-time pattern y.

The length of output vector equals the length of x. Parameter circular determines whether x time-series is assumed to have a circular nature. Assume l_x is the length of time-series x, l_y is the length of short-time pattern y.

If circular equals TRUE then

first element of the output vector corresponds to sample covariance between x[1:l_y] and y,
last element of the output vector corresponds to sample covariance between c(x[l_x], x[1:(l_y - 1)]) and y.

If circular equals FALSE then

first element of the output vector corresponds to sample covariance between x[1:l_y] and y,
the l_x - W + 1-th last element of the output vector corresponds to sample covariance between x[(l_x - l_y + 1):l_x],
last W-1 elements of the output vector are filled with NA.

See runstats.demo(func.name = "RunningCov") for a detailed presentation.

Value

A numeric vector.

Examples

x <- sin(seq(0, 1, length.out = 1000) * 2 * pi * 6)
y <- x[1:100]
out1 <- RunningCov(x, y, circular = TRUE)
out2 <- RunningCov(x, y, circular = FALSE)
plot(out1, type = "l"); points(out2, col = "red")

Fast Running L2 Norm Computation

Description

Computes running L2 norm between between time-series x and short-time pattern y.

Usage

RunningL2Norm(x, y, circular = FALSE)

Arguments

x

A numeric vector.

y

A numeric vector, of equal or shorter length than x.

circular

logical; whether running L2 norm is computed assuming circular nature of x time-series (see Details).

Details

Computes running L2 norm between between time-series x and short-time pattern y. The length of output vector equals the length of x. Parameter circular determines whether x time-series is assumed to have a circular nature. Assume l_x is the length of time-series x, l_y is the length of short-time pattern y.

If circular equals TRUE then

first element of the output vector corresponds to sample L2 norm between x[1:l_y] and y,
last element of the output vector corresponds to sample L2 norm between c(x[l_x], x[1:(l_y - 1)]) and y.

If circular equals FALSE then

first element of the output vector corresponds to sample L2 norm between x[1:l_y] and y,
the l_x - W + 1-th element of the output vector corresponds to sample L2 norm between x[(l_x - l_y + 1):l_x],
last W-1 elements of the output vector are filled with NA.

See runstats.demo(func.name = "RunningL2Norm") for a detailed presentation.

Value

A numeric vector.

Examples

## Ex.1.
x <- sin(seq(0, 1, length.out = 1000) * 2 * pi * 6)
y1 <- x[1:100] + rnorm(100)
y2 <- rnorm(100)
out1 <- RunningL2Norm(x, y1)
out2 <- RunningL2Norm(x, y2)
plot(out1, type = "l"); points(out2, col = "blue")
## Ex.2.
x <- sin(seq(0, 1, length.out = 1000) * 2 * pi * 6)
y <- x[1:100] + rnorm(100)
out1 <- RunningL2Norm(x, y, circular = TRUE)
out2 <- RunningL2Norm(x, y, circular = FALSE)
plot(out1, type = "l"); points(out2, col = "red")

Fast Running Mean Computation

Description

Computes running sample mean of a time-series x in a fixed length window.

Usage

RunningMean(x, W, circular = FALSE)

Arguments

x

A numeric vector.

W

A numeric scalar; length of x window over which sample mean is computed.

circular

Logical; whether running sample mean is computed assuming circular nature of x time-series (see Details).

Details

The length of output vector equals the length of x vector. Parameter circular determines whether x time-series is assumed to have a circular nature. Assume l_x is the length of time-series x, W is a fixed length of x time-series window.

If circular equals TRUE then

first element of the output time-series corresponds to sample mean of x[1:W],
last element of the output time-series corresponds to sample mean of c(x[l_x], x[1:(W - 1)]).

If circular equals FALSE then

first element of the output time-series corresponds to sample mean of x[1:W],
l_x - W + 1-th element of the output time-series corresponds to sample mean of x[(l_x - W + 1):l_x],
last W-1 elements of the output time-series are filled with NA.

See runstats.demo(func.name = "RunningMean") for a detailed presentation.

Value

A numeric vector.

Examples

x <- rnorm(10)
RunningMean(x, 3, circular = FALSE)
RunningMean(x, 3, circular = TRUE)

Fast Running Standard Deviation Computation

Description

Computes running sample standard deviation of a time-series x in a fixed length window.

Usage

RunningSd(x, W, circular = FALSE)

Arguments

x

A numeric vector.

W

A numeric scalar; length of x window over which sample variance is computed.

circular

Logical; whether running sample standard deviation is computed assuming circular nature of x time-series (see Details).

Details

If circular equals TRUE then

first element of the output time-series corresponds to sample standard deviation of x[1:W],
last element of the output time-series corresponds to sample standard deviation of c(x[l_x], x[1:(W - 1)]).

If circular equals FALSE then

first element of the output time-series corresponds to sample standard deviation of x[1:W],
the l_x - W + 1-th element of the output time-series corresponds to sample standard deviation of x[(l_x - W + 1):l_x],
last W-1 elements of the output time-series are filled with NA.

See runstats.demo(func.name = "RunningSd") for a detailed presentation.

Value

A numeric vector.

Examples

x <- rnorm(10)
RunningSd(x, 3, circular = FALSE)
RunningSd(x, 3, circular = FALSE)

Fast Running Variance Computation

Description

Computes running sample variance of a time-series x in a fixed length window.

Usage

RunningVar(x, W, circular = FALSE)

Arguments

x

A numeric vector.

W

A numeric scalar; length of x window over which sample variance is computed.

circular

Logical; whether running sample variance is computed assuming circular nature of x time-series (see Details).

Details

If circular equals TRUE then

first element of the output time-series corresponds to sample variance of x[1:W],
last element of the output time-series corresponds to sample variance of c(x[l_x], x[1:(W - 1)]).

If circular equals FALSE then

first element of the output time-series corresponds to sample variance of x[1:W],
the l_x - W + 1-th element of the output time-series corresponds to sample variance of x[(l_x - W + 1):l_x],
last W-1 elements of the output time-series are filled with NA.

See runstats.demo(func.name = "RunningVar") for a detailed presentation.

Value

A numeric vector.

Examples

x <- rnorm(10)
RunningVar(x, W = 3, circular = FALSE)
RunningVar(x, W = 3, circular = TRUE)

Demo visualization of package functions

Description

Generates demo visualization of output of methods for computing running statistics.

Usage

runstats.demo(func.name = "RunningCov")

Arguments

func.name

Character value; one of the following:

"RunningMean",
"RunningSd",
"RunningVar",
"RunningCov",
"RunningCor",
"RunningL2Norm".

Value

NULL

Examples

## Not run: 
runstats.demo(func.name = "RunningMean")
runstats.demo(func.name = "RunningSd")
runstats.demo(func.name = "RunningVar")
runstats.demo(func.name = "RunningCov")
runstats.demo(func.name = "RunningCor")
runstats.demo(func.name = "RunningL2Norm")

## End(Not run)