The Using of Stochastic Modeling Methods in Criminological Forecasting of Individual Criminal Behavior of Serial Killers

The article considers the stochastic modeling method used in criminology, which is understood as predicting changes in a system with variables ranging randomly with individual probabilities, which allows us to describe the state and probabilistic development of crime, its determinants, individual types, as well as the identity of criminals. The author investigates the possibilities of using this method for criminological purposes, when modeling and predicting the individual criminal behavior of serial killers. The hypothesis and the main conclusions obtained by M. V. Simkin and V. Roychowdhury in stochastic modeling of murders committed by A. R. Chikatilo, in particular, the possibility of using the Cantor`s or «devil's staircase» are checked. Thus, the author analyzed the data on 60 murders committed by M. V. Popkov, also known as the «Angarsk maniac», on the territory of the Irkutsk region in the period from 1992 to 2006. It is suggested that the total number of crimes committed by the «Angarsk maniac» also resembles the «devil's staircase», and the method used can indeed be extended to similar crimes. The author focuses on the fragment of the «Popkov staircase» in the period 1997-1998 as an abnormal surge in the activity of the criminal: an attempt is made to explain the number and frequency of murders by the Kantor staircase method and the pathological drive to kill people diagnosed in the criminal (homicidomania with sadistic elements). Investigation of the actual data on the probabilities of murders committed by M. V. Popkov, obtained by calculation and the theoretical model R. Lange is almost identical, which allows us to model the daily probability of murder as a function of the number of days that have passed since the last crime.

Спектры нормальных мод иерархических ансамблей взаимосвязанных осцилляторов

The family of spectra of normal modes of hierarchical networks of the identical mutually coupled oscillators with different topology of the organization is constructed. It is shown that treelike networks possess a fractal range of normal modes of type " devil's staircase", and with growth of quantity of branching lines and at introduction of additional inter-element couplings of network the quantity of degenerate modes increases. The analysis of influence of topology of networks and forces of communications between its elements on spectral characteristics is provided