Abstract
It is popular belief that the weather is “bad” more frequently on weekends than on other days of the week and this is often perceived to be associated with an increased chance of rain. In fact, the meteorological literature does report some evidence for such human-induced weekly cycles although these findings are not undisputed. To contribute to this discussion, a modern data-driven approach using structured additive regression modelsis applied to a newly available high-quality data set for Austria. The analysis investigates how an ordered response of rain intensities is influenced by a (potential) weekend effect while adjusting for spatio-temporal structure using spatially varying effects of overall level and seasonality patterns. The underlying data are taken from the HOMSTART project which provides daily precipitation quantities over a period of more than 60 years and a dense netof more than 50 meteorological stations all across Austria.
Highlights
Many people have the impression that the weather is “bad” more frequently on weekends when they would be able to enjoy outdoor activities much more than during work days
Baumer and Vogel (2007) report evidence for such weekly patterns in data from 12 German meteorological stations. Such results are not uncontroversial, e.g., the Baumer and Vogel (2007) results have been challenged by Hendricks Franssen (2008) using data for Swiss stations where there was no evidence for weekly patterns if spatial correlations are taken into account
We contribute to the discussion by applying a modern flexible regression model for spatio-temporal data to a novel high-quality precipitation data set for Austria
Summary
Many people have the impression that the weather is “bad” more frequently on weekends when they would be able to enjoy outdoor activities much more than during work days. The statistical model employed assesses the weekend effect while accounting for the inherent temporal and spatial correlations as well as threshold effects in the response by applying a penalized regression approach for an ordered response. It utilizes well-established mixedmodel technology to capture the rather complex and possibly nonlinear relationships in the data (e.g., see Lin and Zhang, 1999 and Kneib and Fahrmeir, 2006)
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
