Dataset of historical GDP per capita estimates
收藏资源简介:
本数据集利用机器学习方法,通过历史人物出生地、死亡地和职业数据,估计了欧洲和北美数十个国家和数百个地区在过去700年的GDP人均值。数据集涵盖了从1300年到2000年的50年间隔的数据,并使用2011年美元购买力平价(PPP)作为单位。这些估计值与历史收入水平的相关性高达90%,并通过与其他经济产出代理变量的比较得到了外部验证。数据集的创建旨在解决历史GDP人均值数据稀缺的问题,为长期经济增长和发展研究提供更多数据支持。
This dataset employs machine learning techniques to estimate per capita GDP for dozens of countries and hundreds of regions across Europe and North America over the past 700 years, utilizing data on historical individuals' birthplaces, places of death, and occupations. The dataset includes observations at 50-year intervals spanning from 1300 to 2000, with all values denominated in 2011 U.S. dollars adjusted for purchasing power parity (PPP). These estimates exhibit a correlation coefficient of up to 90% with observed historical income levels, and have been externally validated via comparisons with other proxy variables for economic output. This dataset was developed to mitigate the scarcity of available historical per capita GDP data, thereby providing enhanced data support for research on long-term economic growth and development.

- 1Augmenting the availability of historical GDP per capita estimates through machine learningCenter for Collective Learning, ANITI, IRIT, Université de Toulouse, Toulouse, France; EcoAustria – Institute for Economic Research, Vienna, Austria; Faculty of Economics, University Ss. Cyril and Methodius, Skopje, North Macedonia; Center for Collective Learning, CIAS, Corvinus University, Budapest, Hungary; Toulouse School of Economics, Université de Toulouse, Toulouse, France · 2025年



