摘要
结合主成分分析(PCA)综合评价理论,构建杠杆率作用于企业创新的智能评价模型,设计基于群体智能优化的杠杆率鲁棒区间优化算法,得到目标创新度下的最优杠杆率区间。进一步采用国泰安数据库提供的中国上市公司数据进行检验,结果显示,所构建的基于RBF神经网络的智能评价模型,高效地评价出随机抽取的12家公司创新度水平;所设计的群优化算法在迭代过程中都能快速地收敛到最优值,且收敛的最小总账面杠杆率和最大总账面杠杆率都出现“中间高两头低”的现象,即目标创新度处于中间段(较大或不大)等级的企业,最小账面杠杆率和最大账面杠杆率都较高,目标创新度处于高段(非常大、很大)等级的企业和目标创新度处于低段(不大、较小)等级的企业,最小账面杠杆率和最大账面杠杆率都较低。由此认为,高段创新度企业,可降低负债,侧重提高资金利用率;中段创新度企业,虽实现技术创新,但还不够规模化或成熟化,因此可增加资金支持力度,来提升创新等级;低段创新度企业,应注重负债规模和技术创新协同调整,提升创新等级,否则过度负债会增加其财务负担。
We first construct an intelligent evaluation model for the impact of leverage on corporate innovation using PCA comprehensive evaluation theory.Then,we design a robust interval optimization algorithm for leverage based on swarm intelligence optimization to determine the optimal leverage interval for the target innovation level.Using data from Chinese listed companies provided by the CSMAR database for testing,the results show that the intelligent evaluation model based on the RBF neural network efficiently evaluates the innovation levels of 12 randomly selected companies.The designed swarm optimization algorithm quickly converges to the optimal value during iterations,with the smallest and largest total book leverage exhibiting a"high in the middle and low on both ends"pattern.This means that for companies with medium(larger or smaller)target innovation levels,both the smallest and largest book leverage are higher,while for companies with high(very large or large)and low(not large or small)target innovation levels,both the smallest and largest book leverage are lower.Based on these findings,the following insights and suggestions are proposed:companies with high innovation levels can reduce debt and focus on improving capital utilization;companies with medium innovation levels,although achieving technological innovation,are not yet scaled or mature enough,so they can increase financial support to enhance their innovation level;companies with low innovation levels should focus on coordinated adjustment of debt scale and technological innovation to improve innovation levels,as excessive debt would otherwise increase their financial burden.
作者
陆晓琴
黄元君
Lu Xiaoqin;Huang Yuanjun(School of Economics and Management,Shanghai Polytechnic University,Shanghai 201209,China;School of Finance,Shanghai Lixin University of Accounting and Finance,Shanghai 201209,China)
出处
《科技管理研究》
2024年第11期47-55,共9页
Science and Technology Management Research
基金
国家社会科学基金一般项目“跨境资本双向动态流动影响宏观杠杆率的机制研究”(22BJL020)
浙江省软科学研究计划项目“杠杆率对企业创新的影响:基于人工智能技术的实证与对策”(2021C35116)。
关键词
杠杆率
企业创新
智能评价模型
鲁棒优化
leverage
corporate innovation
intelligent evaluation model
robust optimization
