Search

By definition, thefts are considered the act of taking away other people's mobile possessions for personal use or for others, affecting crime rates, economic indicators and enabling recent studies to create risk zones in society, contributing to insurance pricing in actuarial methods. This paper analyzes the number of vehicle thefts of 38 locations near Campinas/S˜ao Paulo, Brazil, using a GLARMA(p,q) model with Poisson and Negative Binomial response. The main feature of GLARMA(p,q) is to consider the peculiarities of counting data as high dispersion. As a result, it was possible to verify the adequacy and usefulness of the model for counting data. With specific techniques for estimating time series related to the public security area, patterns can be better understood, revealing relevant information that can be added to decision-making processes to direct public policies. [ABSTRACT FROM AUTHOR]

Metro stations are criminogenic places that radiate and/or absorb crime. In this paper, we assess potential theft transmission in these places and their surrounding areas using police recorded data from the São Paulo metro system in Brazil. The study classifies whether metro stations are crime absorbers, crime radiators, or both, by time of day and day of the week. Time series data were used to test the empirical model (panel vector autoregressive) and Granger tests for temporal precedence. Findings show that 'theft waves' recorded in the metro stations often precede in time those that occur in the stations' surrounding areas. However, on weekdays, during off-peak hours, the opposite might happen: stations absorb some of the crime that spills over from criminogenic surrounding areas into the stations. Since there is evidence of transmission of crime from one environment to another, in both directions, better cooperation between police responsible for security in the surrounding areas and the metro security staff might improve crime prevention initiatives in these transit settings. [ABSTRACT FROM AUTHOR]

This paper analyzes the fluctuations in time series of intentional lethal violent crime rates in Salvador, Bahia, Brazil from 2012 to 2018. The modeling proposed was performed with two main strategies: first, the descriptive of the series under study was defined using the sliding window approach, as determined by autocorrelation from Detrended Fluctuation Analysis (DFA), also with sliding window. Results of this research indicate that the series of DFA exponents, depending on the window (365 or 1000 days) and period, have oscillated between persistent, antipersistent and out of memory. Descriptive analysis identified reduction in the average crime rates from 2013 and a behavioral pattern among other moments. These findings will serve as another methodological proposal for monitoring, control and management of criminal indicators and research on them. • Apply the sliding window methodology in crimes with death. • Present the calculation procedures for the sliding window approach for 365 days (approximately one year) and 1000 days. • Offer a new tool for those interested in evaluating criminal statistics. [ABSTRACT FROM AUTHOR]