An Empirical Investigation Into The Effects Of Replicate Reweighting On Variance Estimates For The Annual Capital Expenditures Survey


Many of the business surveys conducted by the U.S. Census Bureau use a one-stage stratified simple random sample without replacement (SRS-WOR) design. When the sampling unit is an establishment or even an EIN (tax-reporting entity), the sampling rates are generally negligible (e.g., less than 20 percent in all strata). However, if the ultimate sampling unit… (More)

9 Figures and Tables


  • Presentations referencing similar topics