The Effect of CE Sample Sizes on CPI Standard Errors

Jenny Fitzgerald, Harold Gomes, and Imin Hung


The Consumer Price Index (CPI) estimates the change in prices over time of the goods and services U.S. consumers buy for day-to-day living based on price quotes selected from probability samples and aggregation weights derived from the Consumer Expenditure (CE) Survey. Recently, there has been speculation about the percentage of the CPI-U standard error that is due to the variability of its cost weights. Cost weights are the building blocks of the CPI that are simply the product of CPI indexes and CE aggregation weights. In this study, we draw simulated samples of CE reports to produce simulated cost weights; we then use the simulated cost weights to find a functional form of how the CPI-U all U.S. - all items standard error changes as CE sample sizes change using stratified random groups (SRG) variance methodology.