Time Series Forecasting
Predicting Demand, Traffic and Trends With Confidence
- Pages
- 80
- Chapters
- 58
- Level
- Intermediate
- Format
- E-book
About this book
Forecasts guide staffing, stock, budgets and capacity. Done well, they save money; done badly, they create false confidence. This book teaches forecasting that is both useful and honest.
You will learn to explore time series, decompose trend and seasonality, build strong baselines, and apply exponential smoothing, ARIMA-style models, regression with features, and machine-learning approaches. The book gives special attention to evaluation with rolling backtests, prediction intervals, forecasting many series at once, and handling holidays, promotions and sudden shocks.
Worked examples are small enough to check by hand.
What you will learn
- Exploring and decomposing time series
- Baselines that every forecast must beat
- Exponential smoothing and ARIMA-style models
- Machine learning with lags, calendars and external drivers
- Backtesting and choosing forecast metrics
- Prediction intervals and communicating uncertainty
Table of contents
- IntroductionIntroduction
- Chapter 1What Forecasting Is For
- Chapter 2Time Series Data
- Chapter 3Components: Trend, Seasonality and Noise
- Chapter 4Plotting and Exploring
- Chapter 5Decomposition
- Chapter 6Moving Averages
- Chapter 7Stationarity and Differencing
- Chapter 8Preparing Data
- Chapter 9Baselines Every Forecast Must Beat
- Chapter 10Simple Exponential Smoothing
- Chapter 11Trend and Seasonal Smoothing
- Chapter 12ARIMA-Style Models
- Chapter 13Regression With Time Features
- Chapter 14Multiple Seasonalities
- Chapter 15Choosing a Classical Model
- Chapter 16Lag and Window Features
- Chapter 17Machine Learning Models
- Chapter 18Global Models for Many Series
- Chapter 19Hierarchical Forecasting
- Chapter 20Intermittent Demand
- Chapter 21New Products
- Chapter 22Forecasting With External Drivers
- Chapter 23Probabilistic Forecasting
- Chapter 24Point Forecast Metrics
- Chapter 25Scaled Metrics
- Chapter 26Backtesting
- Chapter 27Accuracy by Horizon
- Chapter 28Prediction Intervals
- Chapter 29Quantile Metrics
- Chapter 30From Forecasts to Decisions
- Chapter 31Ensembles
- Chapter 32Holidays and Events
- Chapter 33Promotions and Price
- Chapter 34Shocks and Structural Breaks
- Chapter 35Judgement and Overrides
- Chapter 36Communicating Forecasts
- Chapter 37Forecasting in Code
- Chapter 38Deploying Forecasts
- Chapter 39Monitoring Forecast Accuracy
- Chapter 40Forecastability
- Chapter 41Forecasting Website and System Load
- Chapter 42Forecasting in Healthcare and Public Services
- Chapter 43Energy and Weather-Driven Forecasting
- Chapter 44Financial Forecasting
- Chapter 45Using AI Assistants in Forecasting Work
- Chapter 46Holt's Method, Worked Through
- Chapter 47Anomaly Detection in Time Series
- Chapter 48Nowcasting and Leading Indicators
- Chapter 49Measuring the Effect of an Intervention
- Chapter 50Forecasting With Short Histories
- Chapter 51Data Quality Checks for Time Series
- Chapter 52Case Study: Retail Demand
- Chapter 53Case Study: Contact Centre Staffing
- Chapter 54Case Study: Cash Flow
- Chapter 55Common Mistakes
- Chapter 56Building a Forecasting Practice
- Chapter 57A Project Checklist
- Chapter 58Where to Go Next
- AppendixExercises
- GlossaryGlossary
Who this book is for
Analysts, data scientists, planners and managers who build or use forecasts.
Details
| Format | PDF + web edition |
|---|---|
| Length | 80 pages (approx.) |
| Chapters | 58 |
| Level | Intermediate |
| Language | English |
| Topic | Data |
| Author | Latent Library Editorial |
Keep reading