大数据驱动下新能源汽车政策对关键核心技术创新的动态作用机制研究

Dynamic Mechanism of New Energy Vehicle Policy on Key Technology Innovation Based on Big Data Driven

新能源汽车关键核心技术创新仍存在瓶颈制约,多种政策如何在复杂不确定环境下有效作用亟待探索。基于大数据驱动研究范式,采集12种新能源汽车政策及关键核心技术创新相关多源异构数据,搭建PSR-贝叶斯网络模型,以新能源汽车整车上市企业为研究样本,通过结果预测和原因诊断探究政策对关键核心技术创新的动态作用机制。研究表明:(1)各类政策强度动态变化,呈现由需求面购置拉动,经由供给面推动和环境面压力,向需求面后市场的强度转移趋势;(2)通过结果预测识别各阶段主导作用政策,从初期购置优惠,经由基础设施建设支持与双积分,转向四类政策协同作用为主导;(3)通过原因诊断识别各阶段关键政策瓶颈,科技创新支持、基础设施建设以及需求面非财政措施政策细则需要优化。

New energy vehicle(NEV)market in China has developed rapidly,but there are still bottlenecks in key technology innovation of NEV.It is urgent to explore how various multi-policies play a more effective role in promoting key technological innovation in a complex and uncertain environment.Recent studies mostly use traditional quantitative method to analyze static utility of a certain policy,which is insufficient to analyze the dynamic effect of multiple policies under the complex causal effect.Therefore,on the basis of combing policy informatics and policy complexity theory,this study constructs a dynamic model of policy effect under an uncertain environment based on a big data-driven research paradigm.And then we collect multi-source heterogeneous policy data and build a PSR-Bayesian network model,in order to predict the impact results of multiple policies,and carry out cause diagnosis.This study enriches the quantitative research of policy information under the background of big data,and the research conclusion is of great significance for optimizing the combination structure,the rules and standards of policy measures,and improving the promotion effect of policies on key technological innovations.This study is designed with three parts:The first part is data collection.External embedded multi-source heterogeneous data includes 12 types of policy document data,enterprise innovation process data and innovation result data,and central policy documents from 2014 to 2020 are collected from the official websites of the Ministry of Industry and Information Technology,the Ministry of Science and Technology,the Ministry of Finance,the National Development and Reform Commission and other central ministries and commissions.The financial report data are collected of 16 listed NEV companies from CSMAR database during 2014-2020.The patent data of the above companies are collected from the Incopat patent information platform.The second part is model construction.A policy knowledge discovery model is designed based on PSR theory and Bayesian network model,which clarifies the theoretical logic among policy response,enterprise R&D status and breakthrough pressure of key core technologies.By adjusting the probability of policy nodes and key technology innovation nodes,we explore the dynamic mechanism of policy on key technology innovation.The third part is analysis and discussion.This section is about decision optimization:we firstly adjust the policy response end and the key core technology pressure end respectively.And then we predict the results and diagnose the causes with reference to the initial solution of PSR-Bayesian network,analyze the dynamic changes of policy intensity,and finally find out the leading policies that affect the innovation performance of key technologies and the bottleneck policies that need to be optimized.The study results show that:(1)The policy dynamics change in each stage,the demand-side policy gradually weakens,the supply-side policy gradually strengthens,the environmental regulation policy gradually strengthens,and the environmental support policy shows a trend of first strengthening and then weakening.(2)The results show that there are different policy paths to promote key core technology innovation in each stage.The first stage is the indirect financial market,the second stage is the infrastructure construction,and the third stage is the multi-policy coordination and guidance path.(3)The cause diagnosis shows that there are different policy bottlenecks that restrict the innovation and promotion of key core technologies in each stage.We should identify policies with small positive changes or negative values,and improve the detailed rules and support.This innovation of this paper lies in improving various policy dynamic mechanisms of key technology innovation in a complex and uncertain environment,but it also has some limitations,which are mainly reflected in the fact that the intermediate state after the policy action includes not only R&D investment,but also tax rebate and enterprise income,etc.,so in-depth research can be conducted in the future from the aspects of improving the node information of intermediate state to explore the complex mechanism of policy.