Calculating Confidence Intervals for Regional Economic Impacts of Recreation by Bootstrapping Visitor Expenditures

In this paper I use bootstrap procedures to develop confidence intervals for estimates of total industrial output generated per thousand tourist visits. Mean expenditures from replicated visitor expenditure data included weights to correct for response bias. Impacts were estimated with IMPLAN. Ninety percent interval endpoints were 6 to 16 percent above or below the original sample's point estimate depending on the calculation method. Due to the linearity of input-output a shortcut method that estimates confidence interval endpoints from the distribution of mean expenditure profiles yields nearly identical results