Root Mean Square Error Calculation Matlab Code

Root Mean Square Error (RMSE) is a widely used metric in statistics and machine learning to measure the accuracy of predictive models. This guide explains how to calculate RMSE in MATLAB, including the formula, MATLAB code implementation, and practical examples.

What is Root Mean Square Error?

Root Mean Square Error (RMSE) is a standard way to measure the error of a predictive model by calculating the square root of the average squared differences between predicted and actual values. It provides a measure of how spread out these differences are, with lower values indicating better model performance.

RMSE is particularly useful because it penalizes larger errors more heavily than smaller ones, making it sensitive to outliers. This makes it a good choice for applications where large errors are particularly undesirable.

RMSE Formula

The formula for RMSE is:

RMSE = √(1/n Σ(yᵢ - ȳᵢ)²)

Where:

n = number of observations
yᵢ = actual value
ȳᵢ = predicted value

This formula calculates the square root of the average of the squared differences between the predicted and actual values. The square root ensures the units match the original data, while squaring the errors emphasizes larger differences.

MATLAB Code for RMSE Calculation

MATLAB provides built-in functions to easily calculate RMSE. Here's how to implement it:

To use this code, you'll need MATLAB with the Statistics and Machine Learning Toolbox.

Basic RMSE Calculation

% Define actual and predicted values
actual = [10, 15, 13, 18, 20];
predicted = [12, 14, 13, 17, 19];

% Calculate RMSE
rmse = sqrt(mean((actual - predicted).^2));

% Display result
fprintf('Root Mean Square Error: %.4f\n', rmse);

Using the rmse Function

MATLAB's Statistics and Machine Learning Toolbox includes a built-in rmse function:

% Using the built-in rmse function
rmse_value = rmse(actual, predicted);

% Display result
fprintf('RMSE using built-in function: %.4f\n', rmse_value);

Visualizing RMSE with a Chart

You can visualize the differences between actual and predicted values:

% Create a figure to visualize the differences
figure;
plot(actual, 'b-o', 'DisplayName', 'Actual Values');
hold on;
plot(predicted, 'r--s', 'DisplayName', 'Predicted Values');
xlabel('Observation Number');
ylabel('Value');
title('Actual vs Predicted Values');
legend('show');
grid on;

% Calculate and display RMSE
rmse_value = rmse(actual, predicted);
text(3, max([actual; predicted])*0.9, ...
    sprintf('RMSE: %.4f', rmse_value), ...
    'FontSize', 12, 'Color', 'k', 'BackgroundColor', 'w');

Example Calculation

Let's calculate RMSE for a simple dataset:

Observation	Actual Value	Predicted Value
1	10	12
2	15	14
3	13	13
4	18	17
5	20	19

Using the formula:

RMSE = √[( (10-12)² + (15-14)² + (13-13)² + (18-17)² + (20-19)² ) / 5]

= √[(4 + 1 + 0 + 1 + 1) / 5]

= √(7/5)

= √1.4

= 1.1832

The RMSE for this example is approximately 1.1832, indicating the model's predictions are close to the actual values on average.

Interpreting RMSE Results

Interpreting RMSE requires understanding the context of your data:

RMSE values are in the same units as the original data, making them easy to interpret.
A lower RMSE indicates better model performance.
RMSE should be compared to the range of your data. For example, if your data ranges from 0 to 100, an RMSE of 5 is relatively good, while an RMSE of 20 would indicate poor performance.
RMSE is sensitive to outliers, so it's important to check for extreme values in your data.

When comparing models, always use the same dataset to ensure fair comparison.

FAQ

What is the difference between RMSE and MAE?: RMSE and Mean Absolute Error (MAE) both measure prediction accuracy, but RMSE gives more weight to larger errors because it squares the differences before averaging. MAE treats all errors equally.
When should I use RMSE instead of other metrics?: Use RMSE when you want to penalize larger errors more heavily, such as in financial applications where large errors are particularly undesirable. For more balanced error measurement, consider MAE or MAPE.
How do I know if my RMSE is good?: There's no universal "good" RMSE value - it depends on your specific application and data range. A common approach is to compare your RMSE to the range of your data or to the performance of other models.
Can RMSE be negative?: No, RMSE is always non-negative because it involves squaring the differences before taking the square root. The minimum RMSE is 0, which would occur if all predictions were perfect.
How do I handle missing data when calculating RMSE?: You should exclude observations with missing values from your RMSE calculation. MATLAB's rmse function automatically handles missing data by ignoring them in the calculation.