Introduction to ‘ForecastTB’ package

Demonstration of ‘ForecastTB’ package:

This document demonstates the R package ‘ForecastTB’. It is intended for comparing the performance of forecasting methods. The package assists in developing background, strategies, policies and environment needed for comparison of forecasting methods. A comparison report for the defined framework is produced as an output. Load the package as following:

library(ForecastTB)
#> Registered S3 method overwritten by 'quantmod':
#>   method            from
#>   as.zoo.data.frame zoo

The basic function of the package is prediction_errors(). Following are the parameters considered by this function:

  • data: input time series for testing

  • nval: an integer to decide number of values to predict (default:12)

  • ePara: type of error calculation (RMSE and MAE are default), add an error parameter of your choice in the following manner: ePara = c("errorparametername"), where errorparametername is should be a source/function which returns desired error set. (default:RMSE and MAE)

  • ePara_name: list of names of error parameters passed in order (default:RMSE and MAE)

  • Method: list of locations of function for the proposed prediction method (should be recursive) (default:ARIMA)

  • MethodName: list of names for function for the proposed prediction method in order (default:ARIMA)

  • strats: list of forecasting strategies. Available : recursive and dirRec. (default:recursive)

  • append_: suggests if the function is used to append to another instance. (default:1)

  • dval: last d values of the data to be used for forecasting (default: length of the data)

The prediction_errors() function returns, two slots as output. First slot is output, which provides Error_Parameters, indicating error values for the forecasting methods and error parameters defined in the framework, and Predicted_Values as values forecasted with the same foreasting methods. Further, the second slot is parameters, which returns the parameters used or provided to prediction_errors() function.

a <- prediction_errors(data = nottem)  #`nottem` is a sample dataset in CRAN

a
#> An object of class "prediction_errors"
#> Slot "output":
#> $Error_Parameters
#>            RMSE       MAE      MAPE exec_time
#> ARIMA 2.3400915 1.9329816 4.2156087 0.1036696
#> 
#> $Predicted_Values
#>                    1        2        3        4        5        6        7
#> Test values 39.40000 40.90000 42.40000 47.80000 52.40000 58.00000 60.70000
#> ARIMA       37.41933 37.69716 41.18252 46.29926 52.24804 57.10696 59.71674
#>                    8        9      10       11       12
#> Test values 61.80000 58.20000 46.7000 46.60000 37.80000
#> ARIMA       59.41173 56.38197 51.4756 46.04203 41.52592
#> 
#> 
#> Slot "parameters":
#> $data
#>       Jan  Feb  Mar  Apr  May  Jun  Jul  Aug  Sep  Oct  Nov  Dec
#> 1920 40.6 40.8 44.4 46.7 54.1 58.5 57.7 56.4 54.3 50.5 42.9 39.8
#> 1921 44.2 39.8 45.1 47.0 54.1 58.7 66.3 59.9 57.0 54.2 39.7 42.8
#> 1922 37.5 38.7 39.5 42.1 55.7 57.8 56.8 54.3 54.3 47.1 41.8 41.7
#> 1923 41.8 40.1 42.9 45.8 49.2 52.7 64.2 59.6 54.4 49.2 36.3 37.6
#> 1924 39.3 37.5 38.3 45.5 53.2 57.7 60.8 58.2 56.4 49.8 44.4 43.6
#> 1925 40.0 40.5 40.8 45.1 53.8 59.4 63.5 61.0 53.0 50.0 38.1 36.3
#> 1926 39.2 43.4 43.4 48.9 50.6 56.8 62.5 62.0 57.5 46.7 41.6 39.8
#> 1927 39.4 38.5 45.3 47.1 51.7 55.0 60.4 60.5 54.7 50.3 42.3 35.2
#> 1928 40.8 41.1 42.8 47.3 50.9 56.4 62.2 60.5 55.4 50.2 43.0 37.3
#> 1929 34.8 31.3 41.0 43.9 53.1 56.9 62.5 60.3 59.8 49.2 42.9 41.9
#> 1930 41.6 37.1 41.2 46.9 51.2 60.4 60.1 61.6 57.0 50.9 43.0 38.8
#> 1931 37.1 38.4 38.4 46.5 53.5 58.4 60.6 58.2 53.8 46.6 45.5 40.6
#> 1932 42.4 38.4 40.3 44.6 50.9 57.0 62.1 63.5 56.3 47.3 43.6 41.8
#> 1933 36.2 39.3 44.5 48.7 54.2 60.8 65.5 64.9 60.1 50.2 42.1 35.8
#> 1934 39.4 38.2 40.4 46.9 53.4 59.6 66.5 60.4 59.2 51.2 42.8 45.8
#> 1935 40.0 42.6 43.5 47.1 50.0 60.5 64.6 64.0 56.8 48.6 44.2 36.4
#> 1936 37.3 35.0 44.0 43.9 52.7 58.6 60.0 61.1 58.1 49.6 41.6 41.3
#> 1937 40.8 41.0 38.4 47.4 54.1 58.6 61.4 61.8 56.3 50.9 41.4 37.1
#> 1938 42.1 41.2 47.3 46.6 52.4 59.0 59.6 60.4 57.0 50.7 47.8 39.2
#> 1939 39.4 40.9 42.4 47.8 52.4 58.0 60.7 61.8 58.2 46.7 46.6 37.8
#> 
#> $nval
#> [1] 12
#> 
#> $ePara
#> [1] "RMSE" "MAE"  "MAPE"
#> 
#> $ePara_name
#> [1] "RMSE" "MAE"  "MAPE"
#> 
#> $Method
#> [1] "ARIMA"
#> 
#> $MethodName
#> [1] "ARIMA"
#> 
#> $Strategy
#> [1] "Recursive"
#> 
#> $dval
#> [1] 240

The quick visualization of the object retuned with prediction_errors() function can be done with plot() function as below:

b <- plot(a)

Comparison of multiple methods:

As discussed above, prediction_errors() function evaluates the performance of ARIMA method. In addition, it allows to compare performance of distinct methods along with ARIMA. In following example, two methods (LPSF and PSF) are compared along with the ARIMA. These methods are formatted in the form of a function, which requires data and nval as input parameters and must return the nval number of frecasted values as a vector. In following code, test1() and test2() functions are used for LPSF and PSF methods, respectively.

library(decomposedPSF)
test1 <- function(data, nval){
   return(lpsf(data = data, n.ahead = nval))
}

library(PSF)
test2 <- function(data, nval){
  a <- psf(data = data, cycle = 12)
  b <- predict(object = a, n.ahead = nval)
  return(b)
}

Following code chunk show how user can attach various methods in the prediction_errors() function. In this chunk, the append_ parameter is assigned 1, to appned the new methods (LPSF and PSF) in addition to the default ARIMA method. On contrary, if the append_parameter is assigned 0, only newly added LPSF and PSF nethods would be compared.

a1 <- prediction_errors(data = nottem, nval = 48, 
                        Method = c("test1(data, nval)", "test2(data, nval)"), 
                        MethodName = c("LPSF","PSF"), append_ = 1)
a1@output$Error_Parameters
#>             RMSE        MAE       MAPE  exec_time
#> ARIMA 2.52331560 2.12806408 4.51353777 0.06706047
#> LPSF   2.3915796  1.9361111  4.2386499  0.1389005
#> PSF   2.05773359 1.56168084 3.35323747 0.08567429
b1 <- plot(a1)

Appending new methods:

Consider, another function test3(), which is to be added to an already existing object prediction_errors, eg. a1.

library(forecast)
test3 <- function(data, nval){
  b <- as.numeric(forecast(ets(data), h = nval)$mean)
  return(b)
}

For this purpose, the append_() function can be used as follows:

The append_() function have object, Method, MethodName, ePara and ePara_name parameters, with similar meaning as that of used in prediction_errors() function. Other hidden parameters of the append_() function automatically get synced with the prediction_errors() function.

c1 <- append_(object = a1, Method = c("test3(data,nval)"), MethodName = c('ETS'))
c1@output$Error_Parameters
#>              RMSE         MAE        MAPE   exec_time
#> ARIMA  2.52331560  2.12806408  4.51353777  0.06706047
#> LPSF    2.3915796   1.9361111   4.2386499   0.1389005
#> PSF    2.05773359  1.56168084  3.35323747  0.08567429
#> ETS   38.29743056 36.85216463 73.47667823  0.02986932
d1 <- plot(c1)

Removing methods:

When more than one methods are established in the environment and the user wish to remove one or more of these methods from it, the choose_() function can be used. This function takes a prediction_errors object as input shows all methods established in the environment, and asks the number of methods which the user wants to remove from it.

In the following example, the user supplied 4 as input, which reflects Method 4: ETS, and in response to this, the choose_() function provides a new object with updated method lists.

# > e1 <- choose_(object = c1)
# Following are the methods attached with the object:
#         [,1]    [,2]   [,3]  [,4] 
# Indices "1"     "2"    "3"   "4"  
# Methods "ARIMA" "LPSF" "PSF" "ETS"
#
# Enter the indices of methods to remove:4
#
# > e1@output$Error_Parameters
#            RMSE       MAE exec_time
# ARIMA 2.5233156 2.1280641 0.1963789
# LPSF  2.3915796 1.9361111 0.2990961
# PSF   2.2748736 1.8301389 0.1226711

Adding new Error metrics:

In default scenario, the prediction_errors() function compares forecasting methods in terms of RMSE, MAE and MAPE. In addition, it allows to append multiple new error metrics. The Percent change in variance (PCV) is an another error metric with following definition:

$PCV = \frac{\mid var(Predicted) - var(Observed) \mid}{var(Observed)}$

where var(Predicted) and var(Observed) are variance of predicted and obvserved values. Following chunk code is the function for PCV error metric:

pcv <- function(obs, pred){
  d <- (var(obs) - var(pred)) * 100/ var(obs)
  d <- abs(as.numeric(d))
  return(d)
}

Following chunk code is used to append PCV as a new error metric in existing prediction_errors object.

a1 <- prediction_errors(data = nottem, nval = 48, 
                        Method = c("test1(data, nval)", "test2(data, nval)"), 
                        MethodName = c("LPSF","PSF"), 
                        ePara = "pcv(obs, pred)", ePara_name = 'PCV',
                        append_ = 1)
a1@output$Error_Parameters
#>              RMSE         MAE        MAPE         PCV   exec_time
#> ARIMA  2.52331560  2.12806408  4.51353777 13.75707258  0.05859756
#> LPSF     2.442207    1.925000    4.213824   12.628574    0.132000
#> PSF    2.26573587  1.72187500  3.68420048  0.74537497  0.08170366
b1 <- plot(a1)

A unique plot:

A unique way of showing forecasted values, especially if these are seasonal values, the following function can be used. This plot shows how forecatsed observations are behaving on an increasing number of seasonal time horizons.

plot_circle(a1)
#> `major.tick.percentage` is not used any more, please directly use argument `major.tick.length`.
#> Note: 48 points are out of plotting region in sector 'a', track '1'.

Monte-Carlo strategy:

Monte-Carlo is a popular strategy to compare the performance of forecasting methods, which selects multiple patches of dataset randomly and test performance of forecasting methods and returns the average error values.

The Monte-Carlo strategy ensures an accurate comparison of forecasting methods and avoids the baised results obtained by chance.

This package provides the monte_carlo() function as follows:

The parameters used in this function are:

  • object: output of ‘prediction_errors()’ function

  • size: volume of time series used in Monte Carlo strategy

  • iteration: number of iterations models to be applied

  • fval: a flag to view forecasted values in each iteration (default: 0, don’t view values)

  • figs: a flag to view plots for each iteration (default: 0, don’t view plots)

This function returns:

  • Error values with provided models in each iteration along with the mean values
a1 <- prediction_errors(data = nottem, nval = 48, 
                        Method = c("test1(data, nval)"), 
                        MethodName = c("LPSF"), append_ = 1)
monte_carlo(object = a1, size = 180, iteration = 10)
#>         ARIMA     LPSF
#> 3    4.974178 5.665410
#> 42   2.957357 5.025379
#> 17   3.377959 5.496982
#> 30   3.477685 5.040425
#> 23   3.190408 4.867142
#> 58   2.534576 5.303882
#> 4    5.016741 5.547299
#> 20   3.732034 5.002620
#> 28   5.615779 5.567521
#> 21   2.931270 4.907598
#> Mean 3.780799 5.242426

When monte_carlo() function with fval and figs ON flags:

monte_carlo(object = a1, size = 144, iteration = 2, fval = 1, figs = 1)

#> $Error_Parameters
#>         ARIMA     LPSF
#> 71   2.713594 5.428513
#> 61   5.052924 5.650037
#> Mean 3.883259 5.539275
#> 
#> $Predicted_Values
#> $Predicted_Values[[1]]
#>                    1        2        3        4        5        6        7
#> Test values 36.40000 37.30000 35.00000 44.00000 43.90000 52.70000 58.60000
#> ARIMA       38.18867 36.14923 37.24969 41.49754 47.56345 53.84443 58.67937
#> LPSF        42.10000 35.80000 39.40000 38.20000 40.40000 46.90000 53.40000
#>                    8        9       10       11       12       13       14
#> Test values 60.00000 61.10000 58.10000 49.60000 41.60000 41.30000 40.80000
#> ARIMA       60.85179 59.87343 56.09642 50.59348 44.85618 40.39191 38.32931
#> LPSF        59.60000 66.50000 60.40000 59.20000 51.20000 42.80000 45.80000
#>                   15       16       17       18       19       20       21
#> Test values 41.00000 38.40000 47.40000 54.10000 58.60000 61.40000 61.80000
#> ARIMA       39.13466 42.51006 47.49464 52.73598 56.85568 58.81059 58.15541
#> LPSF        40.00000 42.60000 43.50000 47.10000 50.00000 60.50000 64.60000
#>                   22       23      24       25      26       27       28
#> Test values 56.30000 50.90000 41.4000 37.10000 42.1000 41.20000 47.30000
#> ARIMA       55.14089 50.62714 45.8401 42.03989 40.1902 40.71535 43.40575
#> LPSF        64.00000 56.80000 48.6000 42.10000 35.8000 39.40000 38.20000
#>                   29       30       31       32       33       34       35
#> Test values 46.60000 52.40000 59.00000 59.60000 60.40000 57.00000 50.70000
#> ARIMA       47.49195 51.86296 55.36711 57.11446 56.70156 54.30213 50.60407
#> LPSF        40.40000 46.90000 53.40000 59.60000 66.50000 60.40000 59.20000
#>                   36       37       38       39       40       41       42
#> Test values 47.80000 39.20000 39.40000 40.90000 42.40000 47.80000 52.40000
#> ARIMA       46.61394 43.38402 41.73578 42.05209 44.19045 47.53623 51.17777
#> LPSF        51.20000 42.80000 45.80000 40.00000 42.60000 43.50000 47.10000
#>                   43       44       45       46       47       48
#> Test values 58.00000 60.70000 61.80000 58.20000 46.70000 46.60000
#> ARIMA       54.15382 55.70642 55.47289 53.56867 50.54254 47.21997
#> LPSF        50.00000 60.50000 64.60000 64.00000 56.80000 48.60000
#> 
#> $Predicted_Values[[2]]
#>                    1        2        3        4        5        6        7
#> Test values 42.60000 43.50000 47.10000 50.00000 60.50000 64.60000 64.00000
#> ARIMA       39.85137 41.12164 44.72075 49.21257 53.51536 56.46311 57.34043
#> LPSF        36.20000 39.30000 44.50000 48.70000 54.20000 60.80000 65.50000
#>                    8        9       10       11       12       13       14
#> Test values 56.80000 48.60000 44.20000 36.40000 37.30000 35.00000 44.00000
#> ARIMA       55.98365 52.82756 48.76339 44.89103 42.22197 41.41664 42.62273
#> LPSF        64.90000 60.10000 50.20000 42.10000 35.80000 39.40000 38.20000
#>                   15       16       17       18       19      20       21
#> Test values 43.90000 52.70000 58.60000 60.00000 61.10000 58.1000 49.60000
#> ARIMA       45.45402 49.11026 52.60181 55.01631 55.75594 54.6838 52.14402
#> LPSF        40.40000 46.90000 53.40000 59.60000 66.50000 60.4000 59.20000
#>                  22       23       24       25       26       27       28
#> Test values 41.6000 41.30000 40.80000 41.00000 38.40000 47.40000 54.10000
#> ARIMA       48.8548 45.70669 43.52255 42.84353 43.79642 46.07463 49.03363
#> LPSF        51.2000 42.80000 45.80000 36.20000 39.30000 44.50000 48.70000
#>                   29      30       31       32       33       34      35
#> Test values 58.60000 61.4000 61.80000 56.30000 50.90000 41.40000 37.1000
#> ARIMA       51.87203 53.8477 54.47084 53.62409 51.58057 48.91871 46.3596
#> LPSF        54.20000 60.8000 65.50000 64.90000 60.10000 50.20000 42.1000
#>                   36       37      38       39       40       41       42
#> Test values 42.10000 41.20000 47.3000 46.60000 52.40000 59.00000 59.60000
#> ARIMA       44.57256 44.00092 44.7532 46.58614 48.98067 51.28793 52.90428
#> LPSF        35.80000 39.40000 38.2000 40.40000 46.90000 53.40000 59.60000
#>                   43       44       45       46       47       48
#> Test values 60.40000 57.00000 50.70000 47.80000 39.20000 39.40000
#> ARIMA       53.42849 52.76027 51.11626 48.96226 46.88211 45.42019
#> LPSF        66.50000 60.40000 59.20000 51.20000 42.80000 45.80000

Functions in Future Versions:

  • plot.MC()

  • bollinger_plot()

  • New simulation strategies in prediction_errors()