Accommodation statistics: the current issues and an innovation

Abstract

In tourism accommodation statistics, accuracy (variance and bias) is fundamental. However, minimizing both the variance and bias of the estimate is a challenge due to the trade-off between the two. In many countries, tourism accommodation statistics, which popularly developed with the unbiased linear estimator, are less accurate. This study recommends an integration of bootstrap and regression to overcome the challenge. An unbiased linear estimator is used as a benchmark to investigate the integration method’s effectiveness. With realistic accommodation data in Japan, we found that the integration method yields more than two times more accurate results than the unbiased linear estimator. The integration method contributes to an innovation in accommodation statistics in tourism.