What drives intercity venture capital investment? A comparative analysis between multiple linear regression and random forest

Abstract

Abstract Venture capital (VC) significantly contributes to the development of regional economies and fosters innovation. Analyzing the factors that influence VC investments holds key importance. This study employs two methods to ascertain the relative significance of different factors at the city level: the Lindeman, Merenda, and Gold (LMG) approach in multiple linear regression (MLR) and variable importance in random forest (RF) machine learning. The findings reveal that several factors, including economy, finance, innovation, location, and policy, significantly influence VC investments. Both the MLR and RF models highlight the preeminence of economic and financial variables, followed closely by the city’s potential for innovation. Moreover, spatial heterogeneity exists in the importance of these variables. In the economically developed and densely populated eastern regions of China, the financial environment of cities emerges as the most crucial, whereas in the central and western regions, the economy and innovation, respectively, take precedence. This research contributes to a deeper understanding of the distribution of VC investments and offers valuable insights for the development of regional policies.