Statistical modeling
Statistical modeling is the use of mathematical or probabilistic models to represent patterns, relationships, or variability in data for description, inference, or prediction.
Effective Data Analysis
A statistical model is a mathematical representation of relationships among variables and random variation in observed data.
- How it is used
- Deploys mathematical models to explain relationships, predict outcomes, evaluate evidence, and inform stakeholder decisions.
- How deep it goes
- Explains definitions, purposes, fitting, evaluation, iteration, and limitations, while stopping short of detailed model-specific implementation.
- Examples
- Uses drink-order prediction, New York City rat complaints, advertising conversions, linear regression, residuals, R2, AIC, and BIC.
- What you can do after
- Evaluate model performance, refine fitted models, interpret results, and select deliverables that match stakeholder needs.
Read how Effective Data Analysis explains statistical modeling →
Causal AI
Statistical modeling represents data-generating relationships with probabilistic or deterministic mathematical models.
- How it is used
- Uses statistical assumptions to analyze data and extract patterns, while supporting progressively stronger causal reasoning when causal assumptions are added.
- How deep it goes
- Defines models through data-generating processes and parameter alignment, then distinguishes associational, interventional, and counterfactual assumptions.
- Examples
- Uses in-game purchases, side-quest engagement, linear regression, and neural networks to contrast statistical and causal assumptions.
- What you can do after
- Evaluate/understand how parameterized models represent data patterns and how added causal assumptions change interpretation.
Demand Forecasting Best Practices
A statistical model represents relationships in data using probabilistic or statistical assumptions to describe or predict observations.
- How it is used
- Supports supply-chain forecasting by relating historical demand and demand drivers to predicted demand and planning scenarios.
- How deep it goes
- Explains interpretable time-series components and demand-driver effects, while stopping short of machine-learning complexity and accuracy.
- Examples
- Uses level, trend, seasonality, pricing, promotions, weather impacts, and 15% promotion-uplift scenarios as examples.
- What you can do after
- Evaluate forecasts by tracing unusual results to level, trend, seasonality, or demand-driver behavior.
Read how Demand Forecasting Best Practices explains statistical modeling →
Grokking Statistics
Statistical modeling is the process of representing relationships or data-generating processes with mathematical or probabilistic models.
- How it is used
- Guides data science work by connecting modeling choices to pressing organizational problems and measurable impact.
- How deep it goes
- Emphasizes problem definition and technique selection, but stops short of mathematical derivations, fitting procedures, and model diagnostics.
- Examples
- Contrasts a proposed large language model with a quick regular-expression solution to illustrate technique choice.
- What you can do after
- Apply it via problem-first scoping, technique research, and solution design tied to tangible organizational outcomes.
Read how Grokking Statistics explains statistical modeling →
Related concepts
- datasetuses · A dataset is an organized collection of related data items, records, or observations treated as a unit for storage, processing, analysis, or computation.
- linear regressionincludes · Linear regression is a statistical modeling method that estimates a continuous response as a linear combination of predictor variables and fitted coefficients.
- trainingused by · Training is a process of developing or improving a person’s or model’s capabilities through instruction, experience, or adjustment using data.
- classifierincludes · A classifier is a predictive computational model that assigns inputs to one or more discrete classes or categories.
- Bayesian networkincludes · A Bayesian network is a probabilistic graphical model whose directed acyclic graph represents random variables and their conditional dependencies.
- structural causal modelincludes · A structural causal model represents causal relationships among variables through equations that determine each variable from other variables and exogenous factors.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.