A multiple inflated negative binomial hurdle regression model: analysis of the Italians’ tourism behaviour during the Great Recession

We analyse tourism behaviour of Italian residents in the period covering the 2008 Great Recession. Using the Trips of Italian Residents in Italy and Abroad quarterly survey, carried out by the Italian National Institute of Statistics, we investigate whether and how the economic recession has affecte...

Full description

Saved in:
Bibliographic Details
Published in:Statistical methods & applications 2021-10, Vol.30 (4), p.1109-1133
Main Authors: Bocci, Chiara, Grassini, Laura, Rocco, Emilia
Format: Article
Language:English
Subjects:
Binomial distribution
Chemistry and Earth Sciences
Computer Science
Economic forecasting
Economics
Finance
Great Recession
Health Sciences
Humanities
Insurance
Law
Management
Mathematics and Statistics
Medicine
Original Paper
Physics
Regression models
Specific lengths
Statistical Theory and Methods
Statistics
Statistics for Business
Statistics for Engineering
Statistics for Life Sciences
Statistics for Social Sciences
Tourism
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We analyse tourism behaviour of Italian residents in the period covering the 2008 Great Recession. Using the Trips of Italian Residents in Italy and Abroad quarterly survey, carried out by the Italian National Institute of Statistics, we investigate whether and how the economic recession has affected the total number of overnight stays. The response variable is the result of a two-stage decision process: first we choose to take a holiday, then for how long. Moreover, since the number of overnight stays is typically concentrated on specific lengths (week-end, week, fortnight) we observe multiple peculiar spikes in its distribution. To take into account these two distinctive characteristics, we generalise the usual hurdle regression model by specifying a multiple inflated truncated negative binomial distribution for the positive responses. Results show that the economic recession impacted negatively on both components of the decision process and that, by controlling for the inflated nature of the response variable’s distribution, the proposed formulation provides a better representation of the Italians’ tourism behaviour in comparison with non-inflated hurdle models. Given this, we believe that our model can be a useful tool for policy makers who are trying to forecast the effects of new targeted policies to support tourism economy.
ISSN:1618-2510
1613-981X
DOI:10.1007/s10260-020-00542-6