Package 'forecasteR' reference manual

Title:	Time Series Forecast System
Description:	A web application for displaying, analysing and forecasting univariate time series. Includes basic methods such as mean, naïve, seasonal naïve and drift, as well as more complex methods such as Holt-Winters Box,G and Jenkins, G (1976) <doi:10.1111/jtsa.12194> and ARIMA Brockwell, P.J. and R.A.Davis (1991) <doi:10.1007/978-1-4419-0320-4>.
Authors:	Oldemar Rodriguez [aut, cre], Diego Jiménez [aut]
Maintainer:	Oldemar Rodriguez <[email protected]>
License:	GPL (>=2)
Version:	2.0.2
Built:	2024-11-14 04:30:00 UTC
Source:	https://github.com/PROMiDAT/forecasteR

Best parameters arima model

Description

Best parameters arima model

Usage

calibrar.arima(train, test, period, ar = 0:2, es = 0:1)
calibrar.arima(train, test, period, ar = 0:2, es = 0:1)

Arguments

`train`	a ts object (train of a time series).
`test`	a ts object (test of a time series).
`period`	value indicate the period to use.
`ar`	vector of values to test p, d, q of arima model.
`es`	vector of values to test P, D, Q of arima model.

Value

arima model

Author(s)

Diego Jimenez <[email protected]>

Examples

calibrar.arima(AirPassengers[1:132], AirPassengers[133:144], 12, 0:1)

calibrar.arima(AirPassengers[1:132], AirPassengers[133:144], 12, 0:1)

Best parameters HoltWinters model

Description

Best parameters HoltWinters model

Usage

calibrar.HW(train, test, paso = 0.1)
calibrar.HW(train, test, paso = 0.1)

Arguments

`train`	a ts object (train of a time series).
`test`	a ts object (test of a time series).
`paso`	indicates by value to test alpha, beta and gamma.

Value

HoltWinters model

Author(s)

Diego Jimenez <[email protected]>

Examples

calibrar.HW(window(AirPassengers, end = c(1959, 12)), window(AirPassengers, start = 1960), 0.5)

calibrar.HW(window(AirPassengers, end = c(1959, 12)), window(AirPassengers, start = 1960), 0.5)

Periodogram Data.frame

Description

Periodogram Data.frame

Usage

df_periods(x)
df_periods(x)

Arguments

`x`	a ts object.

Value

data.frame

Author(s)

Diego Jimenez <[email protected]>

Examples

df_periods(AirPassengers)

df_periods(AirPassengers)

Data.frame with normal test

Description

Data.frame with normal test

Usage

dfnormal(data)
dfnormal(data)

Arguments

data

a data.frame object only with the numeric columns.

Value

data.frame

Author(s)

Diego Jimenez <[email protected]>

Examples

dfnormal(iris[, -5])

dfnormal(iris[, -5])

Best parameters arima model

Description

Best parameters arima model

Usage

e_acf(x)
e_acf(x)

Arguments

`x`	a ts object.

Value

echarts4r plot

Author(s)

Diego Jimenez <[email protected]>

Examples

e_acf(AirPassengers)

e_acf(AirPassengers)

Decompose plot

Description

Decompose plot

Usage

e_decompose(serie, f = NULL, noms = NULL)
e_decompose(serie, f = NULL, noms = NULL)

Arguments

`serie`	a ts object.
`f`	vector of dates for the time series.
`noms`	vector of names for y axis.

Value

echarts4r plot

Author(s)

Diego Jimenez <[email protected]>

Examples

e_decompose(AirPassengers)

e_decompose(AirPassengers)

Normal plot

Description

Normal plot

Usage

e_histnormal(
  data,
  colorbar = "steelblue",
  colorline = "gray",
  nombres = c("Histograma", "Curva Normal")
)
e_histnormal(
  data,
  colorbar = "steelblue",
  colorline = "gray",
  nombres = c("Histograma", "Curva Normal")
)

Arguments

`data`	a numeric column of a data.frame.
`colorbar`	a color for the bars.
`colorline`	a color for the line.
`nombres`	a character vector of length 2 specifying the titles to use on legend.

Value

echarts4r plot

Author(s)

Diego Jimenez <[email protected]>

Examples

e_histnormal(iris$Sepal.Length)

e_histnormal(iris$Sepal.Length)

Best parameters arima model

Description

Best parameters arima model

Usage

e_pacf(x)
e_pacf(x)

Arguments

`x`	a ts object.

Value

echarts4r plot

Author(s)

Diego Jimenez <[email protected]>

Examples

e_pacf(AirPassengers)

e_pacf(AirPassengers)

Periodogram Plot

Description

Periodogram Plot

Usage

e_periods(x, p = NULL, noms = NULL)
e_periods(x, p = NULL, noms = NULL)

Arguments

`x`	a ts object.
`p`	which important period to plot.
`noms`	vector of lenght 3 to indicate the text to use.

Value

echarts4r plot

Author(s)

Diego Jimenez <[email protected]>

Examples

e_periods(AirPassengers)

e_periods(AirPassengers)

Qplot + Qline

Description

Qplot + Qline

Usage

e_qq(data, colorpoint = "steelblue", colorline = "gray")
e_qq(data, colorpoint = "steelblue", colorline = "gray")

Arguments

`data`	a numeric column of a data.frame.
`colorpoint`	a color for the points.
`colorline`	a color for the line.

Value

echarts4r plot

Author(s)

Diego Jimenez <[email protected]>

Examples

e_qq(iris$Sepal.Length)

e_qq(iris$Sepal.Length)

Tendencia y Estacionalidad

Description

Tendencia y Estacionalidad

Usage

e_tc(x, d = NULL, noms = c("Time Series", "Trend", "Cyclicality"))
e_tc(x, d = NULL, noms = c("Time Series", "Trend", "Cyclicality"))

Arguments

`x`	a ts object.
`d`	a vector of dates to use on axis x (Optional).
`noms`	a vector of 3 to indicate the names to use on legend.

Value

data.frame

Author(s)

Diego Jimenez <[email protected]>

Examples

e_tc(AirPassengers)

e_tc(AirPassengers)

A web application for displaying, analysing and forecasting univariate time series. Includes basic methods such as mean, naïve, seasonal naïve and drift, as well as more complex methods such as Holt-Winters Box,G and Jenkins, G (1976) <doi:10.1111/jtsa.12194> and ARIMA Brockwell, P.J. and R.A.Davis (1991) <doi:10.1007/978-1-4419-0320-4>.

Details

Package:	forecasteR
Type:	Package
Version:	2.0.2
Date:	2023-06-19
License:	GPL (>=2)

Author(s)

Maintainer: Oldemar Rodriguez Rojas <[email protected]>

Oldemar Rodriguez Rojas <[email protected]>
Diego Jiménez Alvarado

Get ts start of a time series

Description

Get ts start of a time series

Usage

get_start(ini, tipo_f, patron)
get_start(ini, tipo_f, patron)

Arguments

`ini`	a Date object.
`tipo_f`	type of the time series ('year', 'month', ..., 'seconds').
`patron`	frequency of time series.

Value

numeric vector of lenght 2

Author(s)

Diego Jimenez <[email protected]>

Examples

get_start(as.Date("2021-06-30"), 'days', 365)

get_start(as.Date("2021-06-30"), 'days', 365)

Error plot for all predictions

Description

Error plot for all predictions

Usage

grafico.errores(errores)
grafico.errores(errores)

Arguments

errores

a data.frame with errors of a time series.

Value

data.frame

Author(s)

Diego Jimenez <[email protected]>

Examples

model <- arima(window(AirPassengers, end = c(1959, 12)))
pred  <- predict(model, 12)
e <- tabla.errores(list(pred$pred), window(AirPassengers, start = 1960))
grafico.errores(e)

model <- arima(window(AirPassengers, end = c(1959, 12)))
pred  <- predict(model, 12)
e <- tabla.errores(list(pred$pred), window(AirPassengers, start = 1960))
grafico.errores(e)

Mean Square Error

Description

Mean Square Error

Usage

MSE(Pred, Real)
MSE(Pred, Real)

Arguments

`Pred`	a ts object (prediction).
`Real`	a ts object (real).

Value

numeric

Author(s)

Diego Jimenez <[email protected]>

Examples

model <- arima(window(AirPassengers, end = c(1959, 12)))
pred  <- predict(model, 12)
MSE(pred$pred, window(AirPassengers, start = 1960))

model <- arima(window(AirPassengers, end = c(1959, 12)))
pred  <- predict(model, 12)
MSE(pred$pred, window(AirPassengers, start = 1960))

Time series forecasts for a keras model.

Description

Time series forecasts for a keras model.

Usage

pred.tskeras(object, h = 1)
pred.tskeras(object, h = 1)

Arguments

`object`	An object from keras.
`h`	Number of periods for forecasting.

Value

Point forecasts as a time series.

Author(s)

Diego Jimenez <[email protected]>

Examples


  library(keras)
  modelo.deep <- keras_model_sequential() %>% 
  layer_lstm(
    units = 10, activation = 'tanh', batch_input_shape = c(1, 12, 1),
    return_sequences = TRUE, stateful = TRUE) %>%
  layer_dense(units = 1) %>%
  compile(loss = 'mse', optimizer = 'adam', metrics = 'mse')
  modelo.deep <- tskeras(AirPassengers, modelo.deep, lag = 12, epochs = 1)
  pred.tskeras(modelo.deep, h = 12)


library(keras)
  modelo.deep <- keras_model_sequential() %>% 
  layer_lstm(
    units = 10, activation = 'tanh', batch_input_shape = c(1, 12, 1),
    return_sequences = TRUE, stateful = TRUE) %>%
  layer_dense(units = 1) %>%
  compile(loss = 'mse', optimizer = 'adam', metrics = 'mse')
  modelo.deep <- tskeras(AirPassengers, modelo.deep, lag = 12, epochs = 1)
  pred.tskeras(modelo.deep, h = 12)

Relative Error

Description

Relative Error

Usage

RE(Pred, Real)
RE(Pred, Real)

Arguments

`Pred`	a ts object (prediction).
`Real`	a ts object (real).

Value

numeric

Author(s)

Diego Jimenez <[email protected]>

Examples

model <- arima(window(AirPassengers, end = c(1959, 12)))
pred  <- predict(model, 12)
RE(pred$pred, window(AirPassengers, start = 1960))

model <- arima(window(AirPassengers, end = c(1959, 12)))
pred  <- predict(model, 12)
RE(pred$pred, window(AirPassengers, start = 1960))

Root Mean Square Error

Description

Root Mean Square Error

Usage

RMSE(Pred, Real)
RMSE(Pred, Real)

Arguments

`Pred`	a ts object (prediction).
`Real`	a ts object (real).

Value

numeric

Author(s)

Diego Jimenez <[email protected]>

Examples

model <- arima(window(AirPassengers, end = c(1959, 12)))
pred  <- predict(model, 12)
RMSE(pred$pred, window(AirPassengers, start = 1960))

model <- arima(window(AirPassengers, end = c(1959, 12)))
pred  <- predict(model, 12)
RMSE(pred$pred, window(AirPassengers, start = 1960))

RSS

Description

RSS

Usage

RSS(Pred, Real)
RSS(Pred, Real)

Arguments

`Pred`	a ts object (prediction).
`Real`	a ts object (real).

Value

numeric

Author(s)

Diego Jimenez <[email protected]>

Examples

model <- arima(window(AirPassengers, end = c(1959, 12)))
pred  <- predict(model, 12)
RSS(pred$pred, window(AirPassengers, start = 1960))

model <- arima(window(AirPassengers, end = c(1959, 12)))
pred  <- predict(model, 12)
RSS(pred$pred, window(AirPassengers, start = 1960))

Run the Shiny Application

Description

Run the Shiny Application

Usage

run_app(...)
run_app(...)

Arguments

...

A series of options to be used inside the app.

Apply rolling to a numeric vector.

Description

Apply rolling to a numeric vector.

Usage

smoothing(v, n)
smoothing(v, n)

Arguments

`v`	a numeric vector.
`n`	integer value specifying the window width.

Value

numeric vector

Author(s)

Diego Jimenez <[email protected]>

Examples

smoothing(AirPassengers, 5)

smoothing(AirPassengers, 5)

Error table for all predictions

Description

Error table for all predictions

Usage

tabla.errores(Preds, Real, nombres = NULL)
tabla.errores(Preds, Real, nombres = NULL)

Arguments

`Preds`	a list of ts objects (prediction).
`Real`	a ts object (real).
`nombres`	names for the data.frame (optional).

Value

data.frame

Author(s)

Diego Jimenez <[email protected]>

Examples

model <- arima(window(AirPassengers, end = c(1959, 12)))
pred  <- predict(model, 12)
tabla.errores(list(pred$pred), window(AirPassengers, start = 1960))

model <- arima(window(AirPassengers, end = c(1959, 12)))
pred  <- predict(model, 12)
tabla.errores(list(pred$pred), window(AirPassengers, start = 1960))

Convert character to dates

Description

Convert character to dates

Usage

text_toDate(f)
text_toDate(f)

Arguments

`f`	a vector of character.

Value

list

Author(s)

Diego Jimenez <[email protected]>

Examples

text_toDate(c("2023 january 27", "2023 january 28"))

text_toDate(c("2023 january 27", "2023 january 28"))

keras model for time series.

Description

keras model for time series.

Usage

tskeras(x, model, lag = 1, batch_size = 1, epochs = 20, verbose = 0)
tskeras(x, model, lag = 1, batch_size = 1, epochs = 20, verbose = 0)

Arguments

`x`	a ts object.
`model`	a keras model.
`lag`	indicates by value to test alpha, beta and gamma.
`batch_size`	indicates by value to test alpha, beta and gamma.
`epochs`	indicates by value to test alpha, beta and gamma.
`verbose`	indicates by value to test alpha, beta and gamma.

Value

keras model

Author(s)

Diego Jimenez <[email protected]>

Examples


  library(keras)
  modelo.deep <- keras_model_sequential() %>% 
  layer_lstm(
    units = 10, activation = 'tanh', batch_input_shape = c(1, 12, 1),
    return_sequences = TRUE, stateful = TRUE) %>%
  layer_dense(units = 1) %>%
  compile(loss = 'mse', optimizer = 'adam', metrics = 'mse')
  tskeras(AirPassengers, modelo.deep, lag = 12, epochs = 1)


library(keras)
  modelo.deep <- keras_model_sequential() %>% 
  layer_lstm(
    units = 10, activation = 'tanh', batch_input_shape = c(1, 12, 1),
    return_sequences = TRUE, stateful = TRUE) %>%
  layer_dense(units = 1) %>%
  compile(loss = 'mse', optimizer = 'adam', metrics = 'mse')
  tskeras(AirPassengers, modelo.deep, lag = 12, epochs = 1)

Filter category variables of a data.frame

Description

Filter category variables of a data.frame

Usage

var.categoricas(data)
var.categoricas(data)

Arguments

data

a data.frame object.

Value

data.frame

Author(s)

Diego Jimenez <[email protected]>

Examples

var.categoricas(iris)

var.categoricas(iris)

Filter numeric variables of a data.frame

Description

Filter numeric variables of a data.frame

Usage

var.numericas(data)
var.numericas(data)

Arguments

data

a data.frame object.

Value

data.frame

Author(s)

Diego Jimenez <[email protected]>

Examples

var.numericas(iris)

var.numericas(iris)

Package 'forecasteR'

Help Index

Best parameters arima model

Description

Usage

Arguments

Value

Author(s)

Examples

Best parameters HoltWinters model

Description

Usage

Arguments

Value

Author(s)

Examples

Periodogram Data.frame

Description

Usage

Arguments

Value

Author(s)

Examples

Data.frame with normal test

Description

Usage

Arguments

Value

Author(s)

Examples

Best parameters arima model

Description

Usage

Arguments

Value

Author(s)

Examples

Decompose plot

Description

Usage

Arguments

Value

Author(s)

Examples

Normal plot

Description

Usage

Arguments

Value

Author(s)

Examples

Best parameters arima model

Description

Usage

Arguments

Value

Author(s)

Examples

Periodogram Plot

Description

Usage

Arguments

Value

Author(s)

Examples

Qplot + Qline

Description

Usage

Arguments

Value

Author(s)

Examples

Tendencia y Estacionalidad

Description

Usage

Arguments

Value

Author(s)

Examples

Time Series Forecast System