Time Series Decomposition: Separating Signal from Noise
Time series decomposition breaks a metric into its core parts: long-term trend, repeating seasonal patterns, and random noise. This helps you understand *why* a metric changed—was it a real shift or just the usual holiday rush?
The mental model
Time series decomposition is a technique for unmixing a data series into its fundamental components. Instead of looking at a raw line chart, you break it down into three separate parts: the underlying long-term trend, the predictable seasonal patterns, and the leftover random noise or remainder. This lets you analyze each component in isolation to understand the drivers of change.
How it works
Decomposition models the observed data (Yt) as a combination of a trend-cycle component (Tt), a seasonal component (St), and a remainder component (Rt). There are two primary ways to combine them.
First, the additive model: Yt = Tt + St + Rt. Here, the components are simply summed. The seasonal effect is assumed to be a constant value added or subtracted from the trend.
Second, the multiplicative model: Yt = Tt St Rt. Here, the components are multiplied. The seasonal effect is a percentage of the trend. Many multiplicative series can be converted to additive ones by taking the logarithm of the data, since log(A * B) = log(A) + log(B).
When to use it
Use an additive model when the magnitude of seasonal fluctuations is stable regardless of the series' overall level. For example, if an ice cream shop always sells 500 more cones per day in summer than in winter, whether it's a slow year or a busy year, the seasonal effect is additive.
When not to use it
Do not use an additive model if the seasonal variation scales with the level of the series. This is a common footgun. If seasonal swings are proportional, a multiplicative model is more appropriate. For example, if e-commerce revenue always jumps 40% in Q4, the absolute dollar value of that jump will be much larger in a high-revenue year. An additive model would fail to capture this scaling effect and produce poor forecasts.
One canonical example
Consider a company's quarterly sales data. The data shows a general upward trend (Tt) as the company grows. It also shows a strong seasonal pattern (St): sales spike every fourth quarter (Q4) due to holidays and dip in Q1. Because the holiday sales spike is best described as a percentage increase (e.g., +40%) over the current trend, its absolute value grows as the company's baseline sales grow. This makes it a classic case for a multiplicative decomposition: Sales = Trend Seasonality Remainder.
Interview question
Which scenario most strongly suggests using a multiplicative time series decomposition model?
- a.The magnitude of seasonal variations increases proportionally with the overall level of the series.Correct
- b.The long-term trend is the most significant component, dwarfing seasonal effects.
- c.The random noise component is consistently larger than both the trend and seasonal components.
- d.The seasonal fluctuations remain constant in their absolute magnitude over time.
Why? this is the answer
A multiplicative model is appropriate when the seasonal effect scales with the series' overall level, meaning its absolute magnitude grows as the trend grows. Option D describes the condition for an additive model, where seasonal effects have a constant absolute impact.
Just read this? Test yourself on what you have been reading.
Read the original → otexts.com
You just looked this up. Could you explain it out loud?
That is the part interviews actually test. Tezvyn takes questions like this one and gives you what the interviewer is really checking, the answer that lands, and the mistake that ends the conversation, in the four minutes before your next meeting.
The iPhone app is on the way
We are building it. Until it lands, nothing here is held back from you: every interview card, your saved cards, streaks and the job board all work in Safari, plus hundreds of free practice quizzes of thirty questions each. Sign in and it all carries over to the app the day it arrives.
Want it as an icon? Tap Share at the bottom of Safari, then Add to Home Screen. It opens full screen and the cards you have read stay available offline.
We are hiring for this. Open roles that interview on time series — each one lists the topics its interview covers.
See open roles