The Basics
- Simple definition: A subset of artificial intelligence where computers learn from data without being explicitly programmed, identifying patterns and making predictions.
- Core idea: Instead of following rules, algorithms find patterns in data and improve with experience.
- Think of it as: Teaching computers to learn by example, similar to how a child learns to recognize cats by seeing many pictures.
What It Actually Means
Machine learning algorithms build models from training data. Supervised learning involves predicting outcomes from labeled data through regression and classification. Unsupervised learning involves finding patterns in unlabeled data through clustering. Reinforcement learning involves learning by trial and error with rewards. In economics, machine learning is used for forecasting, causal inference with caution, text analysis of news and reports, image data such as satellite imagery for economic activity, and high-dimensional prediction with many variables. Machine learning complements econometrics but does not replace it because it excels at prediction but struggles with causality.
Example
Economists use machine learning to predict inflation from news articles, estimate poverty from satellite images of night lights, detect fraud in tax data, and forecast GDP growth from many indicators. Central banks explore machine learning for nowcasting.
Why It Matters (2026)
Machine learning transforms economic analysis through bigger data, better predictions, and new data sources. Understanding it helps evaluate modern research and its applications in policy, business, and finance.
See also
Artificial Intelligence • Big Data • Deep Learning • Predictive Analytics • Econometrics
Read more about this with MASEconomics: