Temporal Regularities in Homicide: Cycles, Seasons, and Autoregression
First published: 1999
Brief summary
Using monthly US homicide data from 1976-1989, finds statistical evidence for seasonality, autoregression, and a genuine longer-term cyclicality in homicide rates, clarifying previously conflicting research on the topic.
Article
Temporal Regularities in Homicide: Cycles, Seasons, and Autoregression is a peer-reviewed journal article published by Journal of Quantitative Criminology (Springer) in 1999. Using monthly US homicide data from 1976-1989, the study finds statistical evidence for seasonality, autoregression, and a genuine longer-term cyclicality in homicide rates, clarifying previously conflicting research on the topic.
Multiple periodicities tested: annual seasonality plus longer-term cyclicality. The data source is Supplementary Homicide Reports, monthly data 1976-1989. This gives the cycle claim a specific numerical and evidential setting rather than presenting periodicity only as a visual impression.
Employing monthly data from the Supplementary Homicide Reports, the authors report evidence for seasonality, autoregression, and cyclicality of homicide, clarifying previous conflicting research on temporal regularities of homicide. The interpretation is strongest when it survives detrending, autocorrelation controls, structural-break tests and comparison with stochastic or random-walk alternatives.
For cycles researchers, the article brings together homicide cycles, crime seasonality, autoregression, temporal criminology. It is relevant to social-cycle research because apparent waves in historical or behavioural data must be separated from seasonality, autocorrelation, structural change and random walks.
Because it is a peer-reviewed journal article, the article is a strong starting point for discussion in the Sociology forum, although its conclusions should still be compared with later replications and updated datasets.
Source details and credits
- Source / publisher: Journal of Quantitative Criminology (Springer)
- Source type: Peer-reviewed journal article
- URL type: WWW
- Credits: Journal of Quantitative Criminology (Springer)
- URL: https://link.springer.com/article/10.1007/BF02221279