The Appearance And Voice Characteristics Of The Executiveresearch On The Impact Of IPO Market Performance

A large number of studies show that the personal characteristics of senior executives have an important impact on the company’s finance,innovation and future development.In the past,the research on the acquired characteristics of senior executives,such as power structure,educational background,professional experience,and personality(such as narcissism and radicalism),was quite sufficient.In recent years,many scholars have also studied the innate characteristics of senior executives,such as gender,beauty,Facial width-to-height ratio(or short f WHR),and voice.In the IPO market,there are also a lot of studies that show that the innate characteristics of executives have a significant impact on the company’s IPO market.However,these studies are limited to such shallow innate characteristics as gender,age,and a single facial height width ratio;Or indirect methods with more subjective research methods,such as scoring senior executives’ facial beauty and personality based on questionnaires,rarely involve systematic research on the impact of senior executives’ facial geometric features,voice,etc.on the IPO market.First of all,this paper uses the literature research method to establish the research index framework of facial geometric proportion features in this paper.Crawler technology is used to crawl the roadshow video data of listed companies before listing,as well as the published text data of senior executives’ personal introduction.AI face recognition technology is used to extract facial feature points of executives from videos,and geometric scale features are calculated based on the coordinates of feature points;Speech analysis technology is used to extract speech features of senior executives;Text analysis is used to extract the characteristics of senior executives’ gender,age and educational background from their personal introduction texts.The subscription rate of online stock issuance is used as a measure of the popularity of the primary market;The rise and fall on the first day of listing are used as the measurement index of IPO discount rate,and the turnover rate on the first day of listing is used as the measurement index of IPO secondary market heat.Secondly,a stochastic forest prediction algorithm model is established based on the three indicators of all features on the IPO market.According to the contribution index of each feature output from the stochastic forest algorithm to the model,facial features and voice features that have a greater impact on the IPO market indicators are mined.Finally,we establish an econometric model to further explore how these characteristics affect the IPO market,and further verify the results of the random forest model through the significance indicators.The research finds that:(1)The results of the random forest algorithm show that the primary market popularity is most affected by the eight ratio characteristics and the length of senior executives’ speeches;The overall level of the eight ratios in the econometric model is also significantly negatively correlated with the primary market heat,while the speech duration is not significant.(2)In the random forest prediction model of IPO discount rate,it is found that among the facial indicators,classicism and golden ratio have the highest contribution rate to the model;Among the voice indicators,pitch and speech duration have a higher contribution to the model;In the measurement model of IPO discount rate,there are two ratios of classical aesthetic indicators that are significantly negatively correlated.Among the 12 golden ratios,there are three significant negative correlations and three significant positive correlations.The speech duration is still not significant in the voice indicator,but the pitch indicator is significantly positively correlated.(3)The variable with the largest contribution of stochastic forest algorithm features in the secondary market turnover rate model is pitch,and the pitch variable in the econometric model is also significantly positively correlated at the significance level of alpha=0.01.(4)Compared with the conclusions of previous studies,the facial proportion values in this paper are consistent with those in previous studies,and the "bass preference" mentioned by predecessors in sound indicators has also been verified in this paper.