CODEWITHSUNDEEP
rlapplysapplymapply
forecaster
forecaster

Reputation: 1159

applying function across multiple list in R

I have a function that I would like called computeMASE to apply to 3 different lists forecast.list,train.list,test.list all of them have common values (ap,wi). I can use the function individually to the list as shown in the code below, but when I use the lapply and apply the function to get the data all at once, I'm unable to do. Please see below for a reproducible example. Please let me know how to solve this.

Many Thanks

library("forecast")

## Forecast Function

for.x <- function(x){
  fc <- forecast(ets(x),h=18)$mean
  return(fc)
}

## MASE Function

computeMASE <- function(forecast,train,test,period){

  # forecast - forecasted values
  # train - data used for forecasting .. used to find scaling factor
  # test - actual data used for finding MASE.. same length as forecast
  # period - in case of seasonal data.. if not, use 1

  forecast <- as.vector(forecast)
  train <- as.vector(train)
  test <- as.vector(test)

  n <- length(train)
  scalingFactor <- sum(abs(train[(period+1):n] - train[1:(n-period)])) / (n-period)

  et <- abs(test-forecast)
  qt <- et/scalingFactor
  meanMASE <- mean(qt)
  return(meanMASE)
}


## Prepare Data

train.list <- list(ap = ts(AirPassengers[1:(length(AirPassengers)-18)],start=start(AirPassengers),frequency=12),
                  wi = ts(wineind[1:(length(wineind)-18)],end=end(wineind),frequency=12))

test.list <- list(ap = ts(AirPassengers[(length(AirPassengers)-17):length(AirPassengers)],end=end(AirPassengers),frequency=12),
                   wi = ts(wineind[(length(wineind)-17):length(wineind)],end=end(wineind),frequency=12))

## Create Forecast

forecast.list <- lapply(train.list,for.x)

## Compute MASE for each list in the forecast

k.ap <- computeMASE(forecast.list$ap,train.list$ap,test.list$ap,12)
k.wi <- computeMASE(forecast.list$wi,train.list$wi,test.list$wi,12)


## How to apply compute MASE to all the elements in the list,? below does not work

mapply(computeMASE(X,Y,Z,12),X=forecast.list,Y=train.list,Z=test.list)

Upvotes: 3

Views: 6153

Answers (1)

A. Webb
A. Webb

Reputation: 26446

The first argument of mapply should be a function. You can "curry" the period argument

mapply(function(x,y,z) computeMASE(x,y,z,12), forecast.list, train.list, test.list)

or, provide it as another argument (with implicit recycling)

mapply(computeMASE, forecast.list, train.list, test.list, 12)

Upvotes: 7

Related Questions