B1767
Title: Inference in a model for count data with application to Industry 4.0: The permutation approach
Authors: Stefano Bonnini - University of Ferrara (Italy) [presenting]
Michela Borghesi - Università degli Studi di Ferrara (Italy)
Abstract: In order to encourage companies to invest in Industry 4.0 technology, public policy incentives play an important role. The number of 4.0 technologies adopted is represented by a count variable and, according to the literature, this variable should be taken into account when the goal is to measure the propensity to Industry 4.0. The proposed solution to the problem of investigating the effectiveness of public policy incentives on the adoption of 4.0 technologies, concerns the application of a regression analysis for count data and a permutation ANOVA based on the combination of the tests on the significance of the single regression coefficients. The power behaviour of the proposed testing method is studied and compared with some competitors, such as Poisson regression and negative binomial regression, through a Monte Carlo simulation study. Finally, the proposed methodology is applied to an original dataset, related to a sample survey carried out in northern Italy, involving a stratified random sample of Italian small and medium enterprises (SMEs). To avoid the possible confounding effect of some factors, such as firm age, firm size, and economic sector of activity, these elements take the role of control variables.