To Explain or to Predict?

Abstract

Statistical modeling is a powerful tool for developing and testing theories by way of causal explanation, prediction, and description. In many disciplines there is near-exclusive use of statistical modeling for causal explanation and the assumption that models with high explanatory power are inherently of high predictive power. Conflation between explanation and prediction is common, yet the distinction must be understood for progressing scientific knowledge. While this distinction has been recognized in the philosophy of science, the statistical literature lacks a thorough discussion of the many differences that arise in the process of modeling for an explanatory versus a predictive goal. The purpose of this article is to clarify the distinction between explanatory and predictive modeling, to discuss its sources, and to reveal the practical implications of the distinction to each step in the modeling process.

Published in at http://dx.doi.org/10.1214/10-STS330 the Statistical Science (http://www.imstat.org/sts/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (1)

• Boosting Algorithms: Regularization, Prediction and Model Fitting

Cited by in corpus (53)

• Hyperparameters and Tuning Strategies for Random Forest
• Interpretable classifiers using rules and Bayesian analysis: Building a better stroke prediction model
• Autonomous discovery in the chemical sciences part I: Progress
• A review of predictive uncertainty estimation with machine learning
• Super ensemble learning for daily streamflow forecasting: Large-scale demonstration and comparison with multiple machine learning algorithms
• Explainable Artificial Intelligence and Machine Learning: A reality rooted perspective
• A Theory Building Study of Enterprise Architecture Practices and Benefits
• A simple model for predicting the hurricane radius of maximum wind from outer size
• Proton charge radius extraction from electron scattering data using dispersively improved chiral effective field theory
• Tree-based local explanations of machine learning model predictions, AraucanaXAI
• Hydrological time series forecasting using simple combinations: Big data testing and investigations on one-year ahead river flow predictability
• Quantification of predictive uncertainty in hydrological modelling by harnessing the wisdom of the crowd: A large-sample experiment at monthly timescale
• Causal inference in drug discovery and development
• Global-scale massive feature extraction from monthly hydroclimatic time series: Statistical characterizations, spatial patterns and hydrological similarity
• Bayesian hierarchical rule modeling for predicting medical conditions
• Explainable AI for Psychological Profiling from Digital Footprints: A Case Study of Big Five Personality Predictions from Spending Data
• Robust extraction of proton charge radius from electron-proton scattering data
• A Guide to Feature Importance Methods for Scientific Inference
• The worst of both worlds: A comparative analysis of errors in learning from data in psychology and machine learning
• The risk of bias in denoising methods
• Scientific Inference With Interpretable Machine Learning: Analyzing Models to Learn About Real-World Phenomena
• Quantification of predictive uncertainty in hydrological modelling by harnessing the wisdom of the crowd: Methodology development and investigation using toy models
• How Analytic Choices Can Affect the Extraction of Electromagnetic Form Factors from Elastic Electron Scattering Cross Section Data
• Deriving a Usage-Independent Software Quality Metric
• Massive feature extraction for explaining and foretelling hydroclimatic time series forecastability at the global scale
• Privacy, ethics, and data access: A case study of the Fragile Families Challenge
• Comparison of machine learning algorithms for merging gridded satellite and earth-observed precipitation data
• Time-series-analysis-based detection of critical transitions in real-world non-autonomous systems
• Causal propensity as an antecedent of entrepreneurial intentions in tourism students
• Some models are useful, but how do we know which ones? Towards a unified Bayesian model taxonomy
• Predicting Acute Kidney Injury at Hospital Re-entry Using High-dimensional Electronic Health Record Data
• Modeling student pathways in a physics bachelor's degree program
• Key predictors for climate policy support and political mobilization: The role of beliefs and preferences
• Causal inference with multi-state models - estimands and estimators of the population-attributable fraction
• A zero-inflated endemic-epidemic model with an application to measles time series in Germany
• A Novel Stochastic Transformer-based Approach for Post-Traumatic Stress Disorder Detection using Audio Recording of Clinical Interviews
• Forks Over Knives: Predictive Inconsistency in Criminal Justice Algorithmic Risk Assessment Tools
• Metafeatures-based Rule-Extraction for Classifiers on Behavioral and Textual Data
• Differentially Private Release of Israel's National Registry of Live Births
• Simulating outcomes of interventions using a multipurpose simulation program based on the Evolutionary Causal Matrices and Markov Chain
• Predictive Energy Management for Battery Electric Vehicles with Hybrid Models
• Statistically equivalent models with different causal structures: An example from physics identity
• Detailed study of a moving average trading rule
• The MELODIC family for simultaneous binary logistic regression in a reduced space
• A review of machine learning concepts and methods for addressing challenges in probabilistic hydrological post-processing and forecasting
• Occupant Plugload Management for Demand Response in Commercial Buildings: Field Experimentation and Statistical Characterization
• Forbidden Knowledge and Specialized Training: A Versatile Solution for the Two Main Sources of Overfitting in Linear Regression
• SVEMnet: An R package for Self-Validated Elastic-Net Ensembles and Multi-Response Optimization in Small-Sample Mixture-Process Experiments
• Simplifying Causality: A Brief Review of Philosophical Views and Definitions with Examples from Economics, Education, Medicine, Policy, Physics and Engineering
• Complex multiple-choice questions are inequitable for low-income and domestic students (and difficult for everyone)
• Prediction can be safely used as a proxy for explanation in causally consistent Bayesian generalized linear models
• "A 6 or a 9?": Ensemble Learning Through the Multiplicity of Performant Models and Explanations
• Co-Developing Causal Graphs with Domain Experts Guided by Weighted FDR-Adjusted p-values