2021 年 51 巻 1 号 p. 87-99
Based on the cross-sectional data of the tourism economy of China in 2017, using complex network theory, we investigated how to establish a spatial weight matrix to represent the spatial effects among tourism economies of China’s provinces. In addition, with different weight matrices, the construction and optimization of the spatial econometric model were discussed. According to the actual railway line network in China, with complex networks theory, we determined the shortest railway distances between any two regions. Under different thresholds, different weight matrices in the joint spatial econometric model were constructed, and then different complex networks created correspondingly. Afterwards, the parameters in the model were estimated via the maximum likelihood estimation method. Combined with the AIC values, an optimal spatial model was achieved. When the distance was 1,800 km, the spatial model had a stronger interpretation ability. In addition, the diameter of the corresponding network was 4, the average shortest path length was 1.549, and the clustering coefficient was 0.742.
JEL Classifications:R12, O14, Q59