plot_prediction: Plot time series (incl. forecasts) for single cross...

In thfuchs/tsRNN: Time Series Forecasting using Recurrent Neural Networks by Keras

Description

E.g. plot result for selected split from cv_arima (by ggplot2)

Usage

plot_prediction(
  data,
  col_date = "index",
  col_value = "value",
  col_group = "key",
  title = NULL,
  size = 0.6,
  alpha = 0.8,
  legend = "bottom",
  scale = NULL,
  PI = FALSE,
  col_pi_high = "hi95",
  col_pi_low = "lo95"
)

Arguments

data	data.frame containing "index" (Date), "value" (numeric) and key (character - "actual" and "predict")
col_date	Date column in data, default to "index"
col_value	Value column in data, default to "value"
col_group	Group column in data, default to "key" ("actual" and "predict") - see example
title	plot title
size	line size
alpha	line alpha value
legend	legend position
scale	NULL or two Date values to scale x-axis
PI	add prediction interval?
col_pi_high	Upper confidence bound in data
col_pi_low	Lower confidence bound in data

Value

ggplot2 object

Examples

plot_prediction(
  data = tsRNN::fc_arima,
  title = "ARIMA Forecast versus actual data"
)

# With Prediction Interval
plot_prediction(
  data = tsRNN::fc_arima,
  title = "ARIMA Forecast with prediction interval",
  PI = TRUE
)

thfuchs/tsRNN documentation built on April 17, 2021, 11:03 p.m.