Confidence Intervals for Welfare Measures with Application to a Problem of Truncated Counts
用正态、泊松和负二项模型估计狩猎旅行需求，并计算马歇尔和希克斯福利测度的90%置信区间，发现截断对估计有显著影响。
Demand for deer hunting trips was estimated using statistical models based on the normal, Poisson, and negative binomial probability laws. Some of the models accounted for existing sampling truncation. Estimates of Marshallian and Hicksian welfare measures are presented, accompanied by 90 percent confidence intervals based on Krinsky and Robb's procedure. For each of the statistical models, the Hicksian measures are found to be very close to the Marshallian measures, with similar confidence intervals. Accounting for the truncation of the dependent variable has a statistically significant effect on the resulting estimates of welfare measures. Copyright 1991 by MIT Press.