03 — Foundations

Decomposition#

Foundations✓ Mathematical
◆ The PatternSeparating signal from noise, season from trend

Decomposition splits a time series into three components: trend (long-term direction), seasonality (repeating patterns), and residual (noise). This reveals the structure hidden in raw data and guides model choice.

yt = Tt + St + Rt   (additive)   |   yt = Tt × St × Rt   (multiplicative)
Additive when seasonal amplitude is constant. Multiplicative when it grows with the level.
// Interactive — STL decomposition
Seasonal Period12
Trend Strength50
# Python — STL decomposition
from statsmodels.tsa.seasonal import STL

stl = STL(series, period=12, robust=True)
result = stl.fit()
result.plot()
# result.trend, result.seasonal, result.resid
Pattern bridge: Decomposition mirrors eigendecomposition in linear algebra — both break a complex object into orthogonal components. In markets, separating trend from noise is the trader’s version of the same problem.
← Previous
Autocorrelation (ACF/PACF)
Open in the full reader, with the topic sidebar →