A statistical model that predicts the likelihood or probability of an event occurring, often used in predictive analytics to forecast customer behavior, responses, or outcomes.

Predictive Analytics:

The use of data, statistical algorithms, and machine learning techniques to identify the likelihood of future outcomes based on historical data patterns.

Likelihood:

The probability or chance of a specific event or outcome occurring, as estimated by a propensity model.

Target Variable:

The variable or outcome that a propensity model aims to predict or forecast, such as customer conversion, click-through rates, or the likelihood of default.

Training Data:

Historical data used to train and develop a propensity model, typically consisting of examples with known outcomes to teach the model patterns and correlations.

Features:

The variables or attributes used in a propensity model to make predictions, representing the input data that influences the likelihood of the target variable.

Binary Outcome:

A scenario where the target variable has only two possible outcomes, often represented as 0 or 1, such as “converted” or “not converted.”

Logistic Regression:

A statistical method commonly used in propensity models for binary outcomes, estimating the probability of an event occurring.

Machine Learning:

The application of artificial intelligence algorithms that enable computer systems to learn and improve from experience without explicit programming.

Algorithm:

A set of rules or procedures designed to perform a specific task, such as predicting outcomes in a propensity model.

Overfitting:

A situation in which a model learns the training data too well, capturing noise or irrelevant patterns that do not generalize well to new data.

Underfitting:

A situation in which a model is too simplistic and fails to capture the underlying patterns in the training data, resulting in poor predictive performance.

Receiver Operating Characteristic (ROC) Curve:

A graphical representation of a propensity model’s ability to distinguish between true positive rates and false positive rates at different threshold settings.

Area Under the Curve (AUC):

A measure of the overall performance of a propensity model based on the ROC curve, indicating the model’s ability to discriminate between positive and negative cases.

Cross-Validation:

A technique used to assess the performance of a propensity model by splitting the data into multiple subsets for training and testing, helping to avoid overfitting.