Absolute Error
Definition of Absolute Error
Absolute Error is the difference between the predicted value and the actual value.
What is Absolute Error used for?
Absolute Error is a measure of the difference between two values, usually the observed or measured value and the true or accepted value. It is expressed as an absolute value without consideration of sign. In other words, it is the magnitude of the difference between two values regardless if they are negative or positive numbers. Absolute Error is used to measure accuracy in many data science and machine learning applications such as predictive algorithms and regression models.
Absolute Error can be calculated by taking the absolute difference (absolute value) between two numbers, which is equivalent to subtracting one number from the other and then taking the absolute value of the result. For example, if a predicted value for a certain data point is 10 and the true value is 12, then the absolute error would be |10-12| = 2.
In addition to measuring accuracy, Absolute Error can also be used to compare different models with different architectures in order to identify which one has higher accuracy. Furthermore, it can help in identifying problems areas where predictions may not be as accurate as desired and can provide insights into how a model might be improved upon for future iterations. The most important thing about Absolute Error however is that it provides a numerical means of measuring how close any given prediction comes to its actual known value.