Chapter 5: Predictive and Prescriptive Analytics

Vocabulary flashcards covering the key terms, techniques, touchpoints, and categories of predictive, prescriptive, descriptive, diagnostic, and autonomous business analytics.

Confirmatory Business Analytics

A category of business analytics comprising predictive and prescriptive analytics that provides probabilistic models to judge the likelihood of a future event or outcome occurring under conditions of uncertainty.

Probabilistic Data

Data that are not known beforehand and are subject to chance.

Predictive Analytics

An analytics type that provides foresight by identifying patterns in historical data and assessing likelihood or probability to answer questions such as 'Will it happen in the future?' and 'Can we forecast what will happen?'

Classification

A predictive analytics technique and application of artificial intelligence that assigns data into predefined categories or classes using a computer model trained on historical data.

Customer Touchpoints

The various ways in which a company engages with its customers, typically categorized into communication, human, physical, and sensory interactions.

Communication Touchpoints

Channels through which a company communicates with its customers, including websites, email, snail mail, and SMS messaging or texting.

Human Touchpoints

Direct interpersonal interactions between a customer and a salesperson or customer service representative.

Physical Touchpoints

Tangible aspects of customer interaction, including physical store locations, products, and packaging.

Sensory Touchpoints

Environmental and atmospheric elements that engage customer senses, including store fragrances, lighting, and layout design.

Purchase Patterns

A customer's repeated or typical behaviors when making purchases, encompassing purchase frequency, quantity purchased, and purchase timing.

Regression Analysis

A predictive analytics method that uses mathematics and statistics to estimate numerical outcomes based on variable inputs, determining which variables matter most and which can be ignored.

Forecasting with Time Series Analysis

A predictive analytics technique that predicts future values based exclusively on past values of the same single variable.

Persistence

The repeatability, continuity, and durability of critical variables—such as earnings, sales, or cash flows—over time.

Base Rate

The probability of an event occurring based on a related historical average, commonly used to gauge the reasonableness of a prediction.

Base Rate Fallacy

An error that occurs when a prediction places too little weight on historical base rates and instead overly favors new or different information.

Prescriptive Analytics

An analytics type that identifies the best possible options and optimal courses of action given constraints or changing conditions to answer what should be done.

What-If Analysis

An analytical approach that evaluates how potential changes in inputs and assumptions affect output or outcome variables.

Marginal/Incremental Analysis

A prescriptive analytics technique typically used in economics, finance, or accounting that determines the change in profit associated with the cost or benefit of the next (or marginal) unit.

Cash Flow Analysis

A prescriptive technique that evaluates future cash flows for potential investments or expenditures, typically using net present value (NPV) or internal rate of return (IRR).

Scenario Analysis

A prescriptive what-if technique that predicts how changing multiple inputs at the same time will affect an outcome variable.

Goal-Seek Analysis

A form of what-if analysis that works backward by establishing a desired final output or result and determining the required input value(s) needed to achieve it.

Sensitivity Analysis

A prescriptive what-if analysis technique that predicts how changing exactly one input will affect an outcome variable.

Optimization

The process of finding the best feasible solution to make something as effective as possible based on an objective function, decision variables, and constraints.

Objective Function

The core component of an optimization model representing the goal or the specific item to be maximized or minimized.

Decision Variables

The unknown quantities in an optimization model that can vary and must be solved for to achieve the objective.

Constraints

The limited resources and boundaries within an optimization model that determine what an organization can and cannot do.

Descriptive Analytics

An analytics type that characterizes, summarizes, and organizes features and properties of data to facilitate understanding of what happened or what is happening.

Diagnostic Analytics

An analytics type that investigates the underlying cause that cannot be identified merely by descriptive data, explaining why past results occurred through techniques like hypothesis testing.

Adaptive and Autonomous Analytics

Analytics performed by machines that mimic human thought processes and leverage self-learning to offer continuous insights from large volumes of data.