CMStatistics 2022: Start Registration
View Submission - CMStatistics
Title: Markov switching stereotype logit models with two latent indicators for longitudinal ordered data Authors:  Roberto Colombi - University of Bergamo (Italy)
Sabrina Giordano - University of Calabria (Italy) [presenting]
Abstract: Longitudinal ordered categorical data are affected by response styles when respondents are asked to evaluate, on Likert scales, items at different time occasions and decide to use only a few of the given options of the rating scale irrespectively of the content of the item. The novelty, in the context of longitudinal ordered categorical data, is in considering simultaneously the temporal dynamics of observable ordered responses and unobservable answering behaviors, possibly influenced by response styles (RS), through a Markov switching logit model with two latent components. One component accommodates serial dependence and respondent's unobserved heterogeneity, and the other component determines the responding attitude (due to RS or no-RS). The dependence of the observable variables on covariates is modelled by a stereotype logit model with parameters varying according to the two latent indicators. Unobserved heterogeneity, serial dependence and tendency to response style are modelled through our approach on real longitudinal data collected by the Bank of Italy.