Variance Estimations for International Price Program Indexes

Te-Ching Chen, Patrick A. Bobbitt, James A. Himelein, Jr, Steven P. Paben, MoonJung Cho, and Lawrence R. Ernst


The International Price Program (IPP) collects data on United States trade with foreign nations and publishes monthly indexes on import and export prices of U.S. merchandise and services. Recently, the IPP evaluated different variance estimation methods such as Taylor Series Linearization, bootstrap, jackknife, and BRR, for their applicability to the IPP. We constructed an artificial universe of monthly price changes for items constructed from 13 years of IPP historical data. We then compared the bias and stability of the variance estimation methods for month-to-month, annual, and long-term price changes by drawing from the universe 1000 samples in various merchandise strata.