Learn what makes time series data fundamentally different, see the components hiding inside any series, check whether those components are stable enough to forecast, and build models that combine a series' own history with external business drivers. Progress from simple baselines through autoregression to ARX and ARIMAX.
Topics and managerial frameworks covered in Forecasting & Time Series:
Students manipulate these interactive visual simulations directly inside the chapter reading before answering gated check-in questions:
Three pre-loaded marketing time series toggleable via buttons: 1. Monthly DTC Revenue (36 months, trend + strong annual seasonality) 2.
Teaches: time series = trend + seasonality + noise; multiplicative is better when seasonal swings scale with the level.
Shows a non-stationary monthly revenue series (36 points with upward trend). User can apply first differencing and/or seasonal differencing via buttons.
Teaches: raw business series are usually non-stationary; differencing removes trend/seasonality; the ADF test confirms stationarity; the "d" in ARIMA corresponds to how many times you difference.
This chapter connects directly to these full MktPraxis analytical apps (with pre-loaded business case scenarios, CSV upload, and reproducible Python export):
Your first time series tool. See what a series is made of, extend it with a rule you can say out loud, then find out whether that rule beats doing nothing at all.
Pre-loaded Business Scenarios: Paid Search Cost-Per-Click · SaaS Active Subscribers · Garden Centre Units Sold · DTC Apparel Revenue with Promotions
Forecast time series with external predictors. Simplified ARIMAX without moving average terms - runs entirely in browser.
Pre-loaded Business Scenarios: Ice Cream Sales & Temperature · E-commerce Revenue & Ad Spend · Influencer Daily Engagement · Political Comedy Instagram — Daily Likes
Forecast sales over time with external predictors like ad spend.
Pre-loaded Business Scenarios: Monthly Sales with Advertising Spend · Instagram Health Supplement Shop · Mobile Game Monthly Signups
Forecast a yes/no outcome over time from what happened last period and planned drivers, tested walk-forward against a same-as-yesterday baseline.
Pre-loaded Business Scenarios: Is a Third Delivery Worth $180 a Week? Kestrel Market · Do Sell-Outs Come in Streaks? Ferngrove Fresh · Should We Stop the Flash Sales? Brightwick Candles · One Model, Two Suppliers? Kestrel Market
Forecast new product adoption curves, model post-takeoff chasm dynamics, evaluate marketing mix shocks, and simulate multi-generation technological substitution.