Research on Innovation Knowledge Spillover Effect of China’s High-Tech Industry R&D-Base on Multidimensional Spatial Weight Matrices
Xiuwu Zhang, Chengkun Liu
Abstract
Xiuwu Zhang, Chengkun Liu
Abstract
Based on the panel data of R&D activities of the provincial high-tech industry in China from 1998 to 2014, this paper adopts the spatial weight matrix of different dimensions including geographical distance, technical distance, economic distance, proximity distance, and human capital distance, to construct a spatial econometric model to analyze the knowledge spillover effects of R&D activities through both local and transnational routes. The results show that in the case of spatial auto-correlation of the dependent variables, the results of the spatial panel model are more accurate and reliable than those obtained by the conventional panel model. The spatial coefficients of the spatial econometric model based on five different spatial weight matrices are all very significant, and there is a clear spatial correlation between the R&D activities of high-tech industries in different regions. Labor input and exports have a positive impact on innovation output, but the introduction of technology will hinder independent innovation in China’s high-tech industry, and the impact of capital investment to innovation output is uncertain, as it closely relates to the set of models. In addition, the space knowledge spillover effect through the local approach is larger than that produced by the transnational route.
OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Based on the panel data of R&D activities of the provincial high-tech industry in China from 1998 to 2014, this paper adopts the spatial weight matrix of different dimensions including geographical distance, technical distance, economic distance, proximity distance, and human capital distance, to construct a spatial econometric model to analyze the knowledge spillover effects of R&D activities through both local and transnational routes. The results show that in the case of spatial auto-correlation of the dependent variables, the results of the spatial panel model are more accurate and reliable than those obtained by the conventional panel model. The spatial coefficients of the spatial econometric model based on five different spatial weight matrices are all very significant, and there is a clear spatial correlation between the R&D activities of high-tech industries in different regions. Labor input and exports have a positive impact on innovation output, but the introduction of technology will hinder independent innovation in China’s high-tech industry, and the impact of capital investment to innovation output is uncertain, as it closely relates to the set of models. In addition, the space knowledge spillover effect through the local approach is larger than that produced by the transnational route.
Key concepts: Panel data, Spillover effect, Econometric model, Spatial econometrics, Construct (python library), Knowledge spillover, Investment (military), Econometrics