Skip to content
Related Articles

Related Articles

How to Calculate MAPE in R

View Discussion
Improve Article
Save Article
  • Last Updated : 25 Apr, 2022

In this article, we are going to see how to calculate MAPE in R Programming Language.

MAPE:

It is an acronym for mean absolute percentage error. MAPE is used to measure the accuracy of a forecast system. The accuracy is measured in terms of percentage. MAPE can be calculated using the following formula:

MAPE = (1 / n) * Σ(|At – Ft| / |At|) * 100

Here,
n: Represents the number of fitted points
At: Represents the actual value,
Ft: Represents the forecast value.
Σ: Symbol used to denote summation

MAPE is quite common and used to forecast errors. It works most efficiently when their extremes don’t exist in the data.

Method 1: Using MLmetrics package

In R, the MLmetrics package provides us the MAPE() function using which we can calculate the MAPE in R. This function has the following syntax:

Syntax: MAPE(Ft, At)

Parameters:

Here,

  • Ft: It represent forecasted values
  • At: It represents actual values

 Example:

R




# Create a dataframe
dataframe <- data.frame(At=c(15, 40, 41, 32,
                             48, 28, 21, 47, 36,
                             37, 11, 14),
                   Ft=c(32, 41, 43, 54, 66, 51,
                        46, 45, 37, 33, 25, 26))
 
# Print the dataframe
dataframe


Output:

 

Now to find out the MAPE of the above data frame, we can use MAPE() function:

R




# Import library
library("MLmetrics")
 
# Create a dataframe
dataframe <- data.frame(At=c(15, 40, 41, 32,
                             48, 28, 21, 47,
                             36, 37, 11, 14),
                   Ft=c(32, 41, 43, 54, 66,
                        51, 46, 45, 37, 33,
                        25, 26))
 
# Compute MAPE
MAPE(dataframe$Ft, dataframe$At)


Output:

 

Hence, the MAPE value comes out to be equal to 54.915 %.

Method 2: Creating a custom function

We can create our own function to determine MAPE. 

Example:

Let us firstly create a data frame that contains two columns. One column holds actual values and the other column holds forecasted values.  

R




# Create a dataframe
dataframe <- data.frame(At=c(15, 40, 41, 32, 48,
                             28, 21, 47, 36, 37,
                             11, 14),
                   Ft=c(32, 41, 43, 54, 66, 51,
                        46, 45, 37, 33, 25, 26))
 
# Print the dataframe
dataframe


Output:

 

Now we need to create a function that computes MAPE for the above-created data frame:

Note that  (1 / n) * Σ(|At – Ft| / |At|) is equivalent to the mean value of :

abs(( dataframe$At-dataframe$Ft) / dataframe$At)

Thus we can mean() function in R. This function has the following syntax:

Syntax: mean(vect, na.rm)

Parameters:

  • vect: It represents the numeric vector
  • na.rm: Boolean value to ignore NA value

Now we can multiply this value by 100 and return the final value from the function.

R




# Custom function that computes
# MAPE of the passed dataframe
calculateMAPE <- function(dataframe) {
         
        result = mean(abs((dataframe$At-dataframe$Ft)
                          /dataframe$At)) * 100
        return(result)
}
 
# Create a dataframe
dataframe <- data.frame(At=c(15, 40, 41, 32,
                             48, 28, 21, 47,
                             36, 37, 11, 14),
                   Ft=c(32, 41, 43, 54, 66,
                        51, 46, 45, 37, 33,
                        25, 26))
 
# calculate MAPE
result = calculateMAPE(dataframe)
 
# Print the result
print(result)


Output:

 

 Hence, the MAPE value comes out to be equal to 54.915 %.


My Personal Notes arrow_drop_up
Recommended Articles
Page :

Start Your Coding Journey Now!