Skip to content
DeepForgeby svx
Intermediatetime-seriesforecastingexponential-smoothingautocorrelation

Time Series Forecasting

29 problems·~12h·4 stages

Data that moves through time, modeled from first principles. You will build differencing, smoothing, autocorrelation, AR/MA simulation, Holt-Winters, and Yule-Walker solvers without a single library. Suited to analysts and ML engineers working with metrics, demand, or finance.

Your progress

0/29 · 0%

Path readiness 0%Easy 0% · Medium 0% · Hard 0%

Continue path · Mean Absolute Change

What you will learn

  • Transform and difference series into a stationary form
  • Forecast with moving averages, exponential smoothing, and Holt-Winters
  • Fit AR and MA structure with Yule-Walker and Durbin-Levinson
  • Diagnose forecasts with autocorrelation, Ljung-Box, and Theil's U

Before you start

1Series Basics

0/9

Look before you model: absolute changes, first differences, expanding means, seasonal naive forecasts, log transforms, error metrics, stationarity, and rolling means.

Stage checkpoint

0/2 to pass

A checkpoint is a short retrieval test drawn from this stage: solve 2 of 3 from memory to mark the stage complete. The stage stays open either way.

Tried this set and did not pass? Record it — the next set swaps in fresh problems from the same stage.

2Smoothing and Trend

0/7

Averaging with memory: simple and double exponential smoothing, Holt forecasts, autocovariance and autocorrelation, partial autocorrelation, and OLS trend slopes.

Stage checkpoint

0/2 to pass

A checkpoint is a short retrieval test drawn from this stage: solve 2 of 3 from memory to mark the stage complete. The stage stays open either way.

Tried this set and did not pass? Record it — the next set swaps in fresh problems from the same stage.

3AR and MA Models

0/7

Model the dependence structure: MA(1) and AR(1) generation, Holt-Winters, Ljung-Box diagnostics, Yule-Walker solves, and Durbin-Levinson recursion.

Stage checkpoint

0/2 to pass

A checkpoint is a short retrieval test drawn from this stage: solve 2 of 3 from memory to mark the stage complete. The stage stays open either way.

Tried this set and did not pass? Record it — the next set swaps in fresh problems from the same stage.

4Diagnostics and Horizons

0/6

Judge and extend forecasts: ARMA one-step forecasts, Theil's U, the Hurst exponent, lead-lag discovery, and Granger-style predictability tests.

Stage checkpoint

0/2 to pass

A checkpoint is a short retrieval test drawn from this stage: solve 2 of 3 from memory to mark the stage complete. The stage stays open either way.

Tried this set and did not pass? Record it — the next set swaps in fresh problems from the same stage.

Capstone

Contest

Close the path with one real artifact from the catalog.

Forecast Lab

Differences, lags, and a Holt-Winters step against the clock.