As an important part of Spatial Statistics,
spatial autocorrelation analysis studies the degree to which objects or activities
are similar to other objects or activities located nearby at a significance level, and is an efficient means to understand the
spatial pattern based on the relation between objects and their neighbours.Using
spatial autocorrelation techniques,the inbound tourism increase in the period of 1996-2004 in China is studied,and the results show that there is no significant global spatial autocorrelation in the data sets of the inbound tourism increase at province level and the patterns are random on the whole except the year of 1997;there are one or two significant LISAs every year,and the patterns in the subregions are clustered or dispersed respectively.The number of these patterns are changed during the period above,but from 2001,the number is two stably,namely a clustered one influenced markedly by Shanghai and a dispersed one governed by Guangdong province(α=0.05,two-tailed).