📈  Simple Baseline Methods

Simple Baseline Methods#

Easy reference forecasts (last value, mean) that stronger models must beat.

Important

✨ AI-generated content. This page was written with the assistance of an AI language model and is provided as a learning aid. Despite careful review, it may still contain mistakes, omissions, or out-of-date information. Whether you are new to the topic, a team lead, or a senior practitioner, treat it as a starting point rather than an authoritative reference: read it critically and independently verify anything you act on (code, commands, figures, and factual claims) against official documentation and primary sources before relying on it.

What they are#

Simple baseline methods are a small family of trivially simple forecasts used to set a statistical baseline before anything complex. The staples: the mean method (forecast the historical average, which smooths away seasonality), the naïve method (carry the last value forward — a random walk), the seasonal naïve (carry the value from one season ago), and the drift method (naïve plus a trend equal to the average historical change).

The drift formula#

\[\hat{y}_{T+h} = y_T + h \cdot \frac{y_T - y_1}{T-1},\]

equivalent to drawing a line through the first and last observations and extrapolating it forward.

Why use them#

They are cheap and transparent, and any sophisticated model must beat them to earn its complexity. Choose by structure — the mean for a flat series, the naïve for a random walk, the seasonal naïve for strong seasonality, drift for a trend — and note that averaging several often improves accuracy. KISS: keep it sophisticatedly simple.


Theme: Signal Processing & Time Series  ·  All terminology



See also

Source article Adapted (context, re-expressed) in our own words from: Simple Baseline Methods (insightful-data-lab.com).

Tags: purpose: reference topic: terminology level: advanced