Blog/Time Series Analysis

Statistics

Time Series Analysis

Trend, seasonality, and forecasting

SK

Skari Team

Skari

July 2026·14 min read

Time Series

A series carries trend and seasonality; models decompose them and extend the pattern into a forecast.

observedforecast

Almost every method in statistics assumes the rows could be shuffled without losing anything. Time series is the family where that's false. Yesterday's value informs today's; last December looks like this December. The order carries the information, and analyzing it as if the rows were independent throws that information away.

The payoff is forecasting — extending an observed pattern into the future, as the dashed continuation above suggests. But before you can forecast, you have to understand what the series is made of.

Note

The signature property is autocorrelation — each point is correlated with the ones before it. It's exactly what ordinary methods assume away, and exactly what time series methods exploit.

Decomposition: The Parts of a Series

Every series is a sum of a few components. Separating them is the first step and often the most illuminating.

ComponentWhat it is
TrendThe long-run direction — up, down, or flat
SeasonalityA repeating cycle — weekly, monthly, yearly
Residual / noiseWhat's left after trend and season are removed

A sales series that looks chaotic often resolves into a steady upward trend, a strong December spike, and a little noise. Trend analysis and seasonal decomposition make that structure explicit.

Stationarity: The Precondition

Many time series models require a stationary series — one whose statistical properties don't drift over time. A series with a trend or growing variance isn't stationary, and must usually be transformed (often by differencing) before modeling.

Watch out

Fitting a model that assumes stationarity to a trending series produces confident nonsense. Check for stationarity first and difference or transform until the series settles.

The Forecasting Models

ModelBest for
Exponential smoothingTrend and seasonality, simple and robust
ARIMAAutocorrelated series after differencing to stationarity
VAR / VECMSeveral series that move together
GARCH familyVolatility that clusters — finance and risk

ARIMA is the classic workhorse — it models a stationary series from its own past values and errors. Exponential smoothing is often just as good and simpler. VAR handles several interacting series at once; GARCH targets changing volatility rather than the level.

The One Rule: Never Shuffle Time

The most damaging mistake in time series is validating like ordinary data. Randomly splitting the rows lets the model peek at the future to predict the past — a leak that makes a useless model look excellent.

  1. 1Always train on the past and validate on the future, never a random split.
  2. 2Use a forward-chaining (rolling) split so every test window comes after its training window.
  3. 3Judge a forecast on data that genuinely came later, the way production will.

Tip

The cross-validation guide covers the time-series split specifically. If you remember one thing about validating a forecast: time only moves forward, and so must your evaluation.

Time Series in the SKARI Statistical Lab

SKARI's Statistical Lab treats time series as its own family — the largest set of methods — with decomposition, stationarity checks, and forecasting built in.

  • Trend analysis and seasonal decomposition
  • Exponential smoothing and ARIMA forecasting
  • VAR / VECM for multiple interacting series
  • GARCH-family models for volatility
  • Time-aware validation that never shuffles the order

Takeaway

You decompose the series, check stationarity, fit the right forecasting model, and validate forward — so the forecast is judged the way the future will actually arrive.

Frequently Asked Questions

ARIMA or exponential smoothing?

Exponential smoothing is simpler and often competitive for trend-and-season data; ARIMA is more flexible for autocorrelated series. Try both and compare on a forward validation.

What is stationarity, quickly?

A series whose mean and variance don't drift over time. Trends and growing spread break it; differencing usually restores it.

Can I use a normal train/test split?

No — never shuffle time. Train on the past and test on the future, or the score is inflated by leakage.

Key Takeaways

Signal

Order

autocorrelation

Split into

Trend+Season

decompose

Needs

Stationary

difference first

Never

Shuffle

time moves forward

Time series analysis treats order as the signal, not a nuisance. Decompose the series, make it stationary, fit the model that matches its structure, and validate forward — and the forecast holds up when the real future arrives.

Takeaway

The rows can't be shuffled — that's the whole point. Respect the arrow of time in both the model and the validation, and the pattern extends honestly into the future.

Predictive Modeling

Forecasting's supervised cousin

Cross-Validation

The time-aware split

Exploratory Data Analysis

Spot trend and season first