The magazine is publishing the source code of the forecasting model so that readers can download it, explore how it works, tweak its parameters and run it themselves.
The Economist has just launched its first-ever statistical forecast of a US presidential election. The model’s projections are updated daily to reflect the latest available data. “Right now, our model thinks Joe Biden is very likely to beat Donald Trump in the electoral college”, the report says based on regression model used.
Forecasting the US presidential poll in November this year by The Economist –a first, uses polling, economic and demographic data across the states to predict results –these are updated every day.
How reliable is the fundamental model? “….impressively reliable in historical back-testing,” says the international weekly newspaper. It ran the predictability model for the period 1948-2016 for historical back-testing and “were surprised”.
The following three data-sets are being used by the magazine for its daily forecast: Modeled popular vote on each day; Chance of winning each state; and Estimated electoral college votes.