Latent Estimation of GDP, GDP per capita, and Population from Historic\n and Contemporary Sources

Christopher J. Fariss, Charles Crabtree, Therese Anders, Zachary Jones, Fridolin Linder, Jonathan N. Markowitz

arXiv (Cornell University) · 2017 · 33 citations · 0 references

DOIFull text

Open access

Concepts

Abstract

The concepts of Gross Domestic Product (GDP), GDP per capita, and population\nare central to the study of political science and economics. However, a growing\nliterature suggests that existing measures of these concepts contain\nconsiderable error or are based on overly simplistic modeling choices. We\naddress these problems by creating a dynamic, three-dimensional latent trait\nmodel, which uses observed information about GDP, GDP per capita, and\npopulation to estimate posterior prediction intervals for each of these\nimportant concepts. By combining historical and contemporary sources of\ninformation, we are able to extend the temporal and spatial coverage of\nexisting datasets for country-year units back to 1500 A.D through 2015 A.D.\nand, because the model makes use of multiple indicators of the underlying\nconcepts, we are able to estimate the relative precision of the different\ncountry-year estimates. Overall, our latent variable model offers a principled\nmethod for incorporating information from different historic and contemporary\ndata sources. It can be expanded or refined as researchers discover new or\nalternative sources of information about these concepts.\n