liveBook · Manning

Mean absolute percentage error

As explained in Outlier Detection in Python

Mean absolute percentage error is the average absolute forecast error expressed as a proportion of the corresponding actual value.

Mean absolute percentage error (MAPE) averages each absolute forecast error relative to its actual value, expressing typical error as a percentage. section 17.2.6

A MAPE of 0.14, or 14%, means forecasts are off by 14% on average. Use it to assess forecast accuracy or compare tested models. section 17.2.6

Apply MAPE within a forecasting-based outlier workflow:

  • Fit forecasts that are reasonably accurate; weak forecasts make deviations difficult to interpret as outliers. section 17.2.6
  • Evaluate across multiple test months, similarly to cross-validation, rather than relying on one test period. section 17.2.6
  • Inspect unusually large absolute or percentage errors as possible outliers, without treating every flagged value as meaningful automatically. section 17.2.6

How Outlier Detection in Python covers it

How deep it goes
Defines the metric, interprets its percentage, implements its formula, and limits outlier use to reasonably strong forecasts.
Examples
Uses synthetic daily data, 60-day forecasts, a 14% interpretation, and a deliberately increased value producing a large forecast error.
What it assumes
Assumes forecasts, actual values, and sufficiently strong predictive models; teaches the metric’s interpretation and calculation directly.
What you can do after
Evaluate forecast accuracy and compare models with a NumPy MAPE calculation, then inspect large errors for possible outliers.
How it connects
Connects time-series forecasting with forecast accuracy, cross-validation, absolute error, squared error, and outlier detection.
Cover of Outlier Detection in Python
liveBookmanning.com
Read Outlier Detection in Python on liveBook Get the book at manning.com

Also explained in

Compare how 3 books explain mean absolute percentage error →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Outlier Detection in Python. The section links point to the parts of the book it draws on.