forecast_accuracy: Calculate forecast accuracy

In tylerJPike/OOS: Out-of-Sample Time Series Forecasting

Description

A function to calculate various loss functions, including MSE, RMSE, MAE, and MAPE.

Usage

forecast_accuracy(Data)

Arguments

Data	data.frame: data frame of forecasts, model names, and dates

Value

data.frame of numeric error results

Examples

 # simple time series
 A = c(1:100) + rnorm(100)
 date = seq.Date(from = as.Date('2000-01-01'), by = 'month', length.out = 100)
 Data = data.frame(date = date, A)

 # run forecast_univariate
 forecast.uni =
   forecast_univariate(
     Data = Data,
     forecast.dates = tail(Data$date,10),
     method = c('naive','auto.arima', 'ets'),
     horizon = 1,
     recursive = FALSE,
     freq = 'month')

 forecasts =
   dplyr::left_join(
     forecast.uni,
     data.frame(date, observed = A),
     by = 'date'
   )

 # forecast accuracy
 forecast.accuracy = forecast_accuracy(forecasts)

tylerJPike/OOS documentation built on March 20, 2021, 3:20 a.m.