The leading indicators model, developed by Robert Erikson and Christopher Wlezien, predicts the two-party popular vote based on two variables: an index of leading economic indicators and trial-heat polls over the course of the campaign. The model’s vote equation reads as:
V = A + b1
Cumulative LEI Growth 13 + b2
|Table 1: Overview of variables used in the leading indicators model|
||Summed weighted growth in leading economic indicators through quarter 13 of the election cycle, with each quarter weighted 0.8 times the following quarter|
||Incumbent party’s candidate two-party support in polls in month X|
|V||Incumbent share of the two-party presidential vote|
As shown in the table below, the model has published forecasts of the two-party vote in US presidential elections since 1992. Across these eight elections, the model forecasts, published at least two months before the election, was lower than two percentage points.
Erikson, R. S. & Wlezien, C. (2016). Forecasting the Presidential Vote with Leading Economic Indicators and the Polls. PS: Political Science & Politics, 49(4), 669-672.