This page explains the data sources, assumptions and calculation methodology used throughout Horizon Works Lab.
The analysis is based on long-term historical asset class returns compiled by Professor Aswath Damodaran at New York University Stern School of Business.
The dataset covers major investable asset classes including stocks, bonds, treasury bills and gold. I chose investable asset classes rather than theoretical factor portfolios because they are easier to relate to and compare with real-world investment decisions.
Rather than showing the experience of an investor who started in a single year, the tool evaluates every possible starting year available in the dataset.
For example, a 20-year analysis examines every historical 20-year investment period and displays the full distribution of outcomes. This approach helps illustrate the range of experiences investors could have encountered across different market environments.
All results are adjusted for inflation using U.S. Consumer Price Index (CPI) data.
This allows outcomes to be viewed in terms of purchasing power rather than nominal returns, making historical comparisons more meaningful across long time periods.
Portfolios are rebalanced annually to their target weights.
The objective is to isolate the impact of asset allocation decisions while maintaining a simple and transparent methodology.
The goal is not to predict the future, but to help investors build intuition about the relationship between risk, time horizon and portfolio construction.
Huge thanks to Aswath Damodaran for making long-term historical market data publicly available to investors, students and researchers around the world.