The impact of innovation on companies’ performance: an entropy-based analysis of the STAR market segment of the Italian Stock Exchange

Abstract

This paper analyzes the links between innovation pre-crisis and performance during crisis for firms listed in the STAR market segment of the Italian Stock Exchange. The aim is to shed light on the relevant parameters for explaining the relationship between innovation and performance (in terms of growth, profitability and productivity), with special focus on innovation type, innovation level and business size. An empirical analysis based on the perspective of information theory is carried out by introducing suitable classes of entropy measures on aggregated quantities. In so doing, results are not biased by the probability estimates of the empirical data, in that entropy-type measures involve directly empirical distributions, so no hypothesis test is required and no best fit of known probability distribution of data is considered. The study shows that the type of innovation is significantly linked with the performance: there is evidence that process innovation initiatives imply better performance.

Exploring how innovation strategies at time of crisis influence performance: a cluster analysis perspective

“…Specifically, we take into account aggregated indicators for growth, profitability and productivity. Bartolacci et al (2015) employ the same sample analyzed here and discuss the effects of innovation on performance through a new class of entropy measures. In particular, the quoted paper seeks the similarities between companies in terms of the disorder generated by their classifications.…”

Relationship between Entropy, Corporate Entrepreneurship and Organizational Capabilities in Romanian Medium Sized Enterprises

“…The non-significance of the company age control variable in our regression analyses suggests that a survivor bias is unlikely to have biased our results [48]. Finally, complex data relationships such as the expected interaction effects suggest that common method bias should not have led to the significant findings of interaction effects, as quadratic and interaction effects can be severely deflated through common method variance and would be more difficult to defect if common method bias existed [49]. Respondents would normally not be capable of guessing the interaction hypothesis [43].…”

Inteligência competitiva, inovação e performance: proposta de modelo teórico

“…Oliver et al 2015Pesquisaram organizações espanholas e os resultados indicaram que a introdução conjunta da gestão de inovações com as inovações tecnológicas melhora a performance graças à criação de capacidades de inovação complexas. Bartolacci et al (2015) Em uma pesquisa realizada em organizações de um setor específico, os resultados evidenciaram que os decisores políticos podem planejar e implementar estratégias de inovação eficazes para melhorar a performance. Nunes e Lopes…”