liveBook · Manning

Forecasting

As explained in Outlier Detection in Python

The process of predicting future values or events from historical or current data.

Forecasting predicts future values from historical or current time-series data. For outlier detection, a point becomes suspicious when it differs substantially from what the model expected. section 17.2.6

A practical forecasting workflow includes:

  • Model recent values, trends, ruptures, and cycles with methods such as ARIMA or Exponential Smoothing. section 17.2.6
  • Convert forecast error into an outlier score, preferably relative to normal errors using the median and median absolute deviation. section 17.2.6
  • Compare each observation with a confidence range when the forecasting tool provides one, while accounting for wider uncertainty farther into the future. section 17.3.3

Detecting an anomalous forecast error

A model forecasts a value of 50 for a time point, but the actual value is 240, producing an error of 190. If ordinary errors are around 10 or 20, the observation is unusually far from expectation. section 17.2.6

How Outlier Detection in Python covers it

How deep it goes
Explains forecast errors, robust scoring, confidence ranges, model fitting, evaluation, and horizon limitations without deriving forecasting theory.
Examples
Demonstrates synthetic weekly data, ARIMA and Random Forest forecasts, an injected outlier, MAPE evaluation, and UPS Twitter volume with Prophet.
What it assumes
Assumes tabular time-series data and basic model fitting, while introducing lag features, forecast errors, confidence ranges, and temporal evaluation concerns.
What you can do after
Apply forecasting with Darts, scikit-learn, or Prophet to estimate values, measure errors, and flag unusually large deviations.
How it connects
Connects time-series analysis with outlier detection, ARIMA, Exponential Smoothing, Random Forest, Prophet, trends, cycles, and change-point 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 forecasting →

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.