plot.visitation_forecast_ensemble {VisitorCounts}

visitation_forecast_ensemble Plot Methods

Description

Methods for plotting objects of the class "visitation_forecast_ensemble".

Usage

## S3 method for class 'visitation_forecast_ensemble'
plot(
  x,
  difference = FALSE,
  log_outputs = FALSE,
  plot_cumsum = FALSE,
  plot_percent_change = FALSE,
  actual_visitation = NULL,
  actual_visitation_label = "Actual",
  xlab = "Time",
  ylab = "Fitted Value",
  pred_colors = c("#ff6361", "#58508d", "#bc5090", "#003f5c"),
  actual_color = "#ffa600",
  size = 1.5,
  main = "Forecasts for Visitation Model",
  plot_points = FALSE,
  ...
)

Arguments

x

An object of class "visitation_forecast_ensemble".

difference

A Boolean specifying whether to plot the original fit or differenced series. The default option is FALSE, in which case, the series is not differenced.

log_outputs

A Boolean specifying whether to plot the outputs in standard scale or the logged outputs

plot_cumsum

A Boolean specifying whether to plot the cumsum of the outputs

plot_percent_change

A Boolean specifying whether to plot the percent change of the outputs

actual_visitation

A time series object representing the actual visitation in the standard scale

actual_visitation_label

A Label for the actual visitation in the legend

xlab

A String to overwrite the x label of the graph

ylab

A String to overwrite the y label of the graph

pred_colors

An array of strings specifying the colors for the predicted outputs.

actual_color

A String to specify the color of the line for actual visitation

size

A number representing the size of the line to plot.

main

A String to overwrite the main label of the graph

plot_points

A Boolean specifying whether to plot the individual points of visitation

...

Additional arguments.

Examples

data("park_visitation")
data("flickr_userdays")
YELL_data <- park_visitation[park_visitation$park == "YELL",]
YELL_data

YELL_PUD <- YELL_data$pud #PUD Data
YELL_NPS <- YELL_data$nps #NPS Data

#The YELL data from 2005 through 2016 are used as the training data.
YELL_data[1:144,]$date
YELL_PUD.train <- ts(YELL_PUD[1:144], start = c(2005,1), end = c(2016,12), freq = 12)
YELL_NPS.train <- ts(YELL_NPS[1:144], start = c(2005,1), end = c(2016,12), freq = 12)


YELL_NPS.test <- ts(YELL_NPS[144:length(YELL_NPS)],
                   start = c(2016,12), end = c(2017,12), freq = 12)

#Construct models without linear trend (with or without OSC).
YELL_model.without_trend <- visitation_model(onsite_usage = YELL_PUD.train,
                                             ref_series = YELL_NPS.train,
                                             parameter_estimates = "joint", trend = "none")

YELL_model.without_trend_and_NPS <- visitation_model(onsite_usage = YELL_PUD.train,
                                                     ref_series = NULL)

YELL_pred.without_trend <- predict(YELL_model.without_trend,n_ahead = 12)
YELL_pred.without_trend_and_NPS <- predict(YELL_model.without_trend_and_NPS, n_ahead = 12)
YELL_NPS.test



forecast_ensemble <- new_visitation_forecast_ensemble(list(YELL_pred.without_trend ,
YELL_pred.without_trend_and_NPS), list("Without Trend", "Without Trend and NPS"))

#Plot the forecasts and actual percent changes for 2017.
plot(forecast_ensemble, actual_visitation = YELL_NPS.test,
     ylab = "Percent Change in Monthly Visitation", main = "Forecast vs. 
Actual Monthly Percent Change in Visitation in 2017", plot_percent_change = TRUE)


#Plot the cumulative forecasts and actual cumulative percent changes for 2017.
plot(forecast_ensemble, actual_visitation = YELL_NPS.test,
     ylab = "Cumulative Percent Change in Monthly Visitation", 
main = "Forecast vs. Actual Monthly Percent Change in Visitation in 2017", plot_cumsum = TRUE)