创新投入对大数据企业绩效的影响及其门槛效应

Impact of innovation input on big data enterprise performanceand its threshold effect

作为一项风险投资活动,创新投入是大数据企业开展创新活动的重要支撑,但其效应具有较高的不确定性。对创新投入与大数据企业绩效的实证研究成果很有限,迄今尚未有学者关注非线性影响。探讨创新投入对大数据企业绩效的影响及其门槛效应,对于大数据企业通过创新投入提升绩效具有重要的理论和现实意义。为此,从理论层面梳理创新投入对大数据企业绩效的影响,在此基础上以2013-2019年沪深A股上市大数据企业为研究样本,构建基准回归模型、调节效应模型及门槛效应模型,实证检验创新投入对企业绩效的影响,并考察了技术积累的调节效应及创新投入与企业绩效之间存在的非线性关系。研究发现,创新投入有助于提升大数据企业绩效,且影响存在一定的滞后性。这一结论在经过引入工具变量、混合回归及替换被解释变量等一系列内生性检验和稳健性检验之后依然成立。调节效应检验表明,技术积累对创新投入与企业绩效关系具有负向调节效应。门槛效应检验表明,创新投入对企业绩效的影响存在非线性关系,需要控制创新投入的比例。因此,大数据企业应制定可持续的创新投入规划,合理配置内部研发资源,重视创新人才的引进和培养,并寻求与研究机构的合作;加强创新管理,提高企业探索性创新能力;科学制定计划,建立内部预警机制,防止过度创新投入抑制企业绩效。

As a venture capital activity,innovation investment is an important support for big data enterprises to carry out innovation activities,but its effects have high uncertainty.The empirical research results on innovation investment and big data enterprise performance are still limited,and so far,there has been no attention to nonlinear effects.Exploring the impact of innovation investment on the performance of big data enterprises and its threshold effect has important theoretical and practical significance for big data enterprises to improve their performance through innovation investment.Therefore,the paper theoretically examines the impact of innovation investment on the performance of big data enterprises.Based on this,benchmark regression models,moderating effect models,and threshold effect models are constructed using big data enterprises listed on the Shanghai and Shenzhen A-shares from 2013 to 2019 as research samples.Empirical tests are conducted to examine the impact of innovation investment on enterprise performance,and to examine the moderating effect of technology accumulation and the nonlinear relationship between innovation investment and enterprise performance.Research has found that innovation investment can help improve the performance of big data enterprises,and the impact has a certain lag.This conclusion is still valid after a series of endogenous tests and robustness tests,such as the introduction of instrumental variables,mixed regression and substitution of explained variables.The moderating effect test shows that technology accumulation has a negative moderating effect on the relationship between innovation investment and enterprise performance.The threshold effect test indicates that there is a non-linear relationship between the impact of innovation investment on enterprise performance,and it is necessary to control the proportion of innovation investment.Consequently,big data enterprises should devise sustainable innovation investment strategies,allocate R&D resources efficiently,emphasize the recruitment and development of innovative talent,and collaborate with research institutions.They should also strengthen innovation management,enhance exploratory innovation capabilities,develop scientific plans,establish internal warning mechanisms,and prevent excessive innovation investment from hindering performance.