VAR model with exogenous variable doesn't work in Shiny

问题

Let's consider a VAR model with an exogenous variable to distinguish two periods. This model works perfectly as follows:

library(shiny)
library(vars)

#--- Create Exogenous Variable 'periods'
data(Canada)
canTS <- Canada
periods <- as.matrix(data.frame(period=ifelse(index(canTS)>1996, 1, 0 ) ) )

#--- Fit the Model
fit1 <- VAR(Canada, p = 2, type = "none", exogen=periods)
coef(fit1)[[1]]

#-- Make Prediction
period2 <- as.matrix(data.frame(period = rep(1, 12)) ) # Future Exogen Values = 1 
predict(fit1, n.ahead=12, dumvar=period2)

rm(list=ls(all=TRUE)) # remove objects

Problem appears when the Shiny app with slider input (which specifies a forecasting horizon) is used. We can create a simple Shiny app as:

ui <- fluidPage(
 sidebarLayout(
  sidebarPanel(
   sliderInput(inputId = "nAhead",
    label = "Forecast Period",
    min = 1,
    max = 12,
    value = 6)
  ),
  mainPanel(
   verbatimTextOutput("Model")
  )
 )
)

server <- function(input, output) {
 #--- Create Exogenous Variable 'periods'
 data(Canada)
 canTS <- Canada
 periods <- as.matrix(data.frame(period=ifelse(index(canTS)>1996, 1, 0 ) ) ) 

 #--- Fit the Model
 fit1 <- VAR(Canada, p = 2, type = "none", exogen=periods)

 #-- Make Prediction
 output$Model <- renderPrint({
 periods2 <- as.matrix(data.frame(period = rep(1, input$nAhead)) ) # forecasting window
 predict(fit1, n.ahead=input$nAhead, dumvar=periods2) 
 })
}

shinyApp(ui, server)

I receive following error message: "object periods not found". I don't understand why the predict function does not accept a new data matrix supplied by dumvar. Any idea how to make this working? Thank you.

回答1:

After the line

periods <- as.matrix(data.frame(period=ifelse(index(canTS)>1996, 1, 0 ) ) )

just add the following line

assign("periods",periods, envir = .GlobalEnv)

Then it works

来源：https://stackoverflow.com/questions/50531633/var-model-with-exogenous-variable-doesnt-work-in-shiny

标签

Shiny

prediction