Abstract: 21 prefecture-level cities in Guangdong province are taken as a sample, and spatial autocorrelation and spatial autoregressive models are used to demonstrate the differences in venture capital among cities and to explore the influencing factorsThe results show that the spatial autocorrelation analysis method can solve the spatial correlation problem among regions, and better depict the spatial distribution pattern of variables and reflect the spatial heterogeneity among regionsSome other conclusions are drawn as followsFirst, the venture capital in Guangdong province is concentrated in the Pearl River Delta region with Guangzhou and Shenzhen as its coreSecond, there is a positive spatial correlation of the venture capital in each prefecture-level city, and the change keeps more stable situationThird, scientific and technological innovation, the government’s investment in science and technology, human capital, economic development and transport infrastructure are the main factors affecting the venture capital in Guangdong province.

Key words:  venture capital, spatial autocorrelation, spatial lag model, Guangdong province