Complex analysis of combat in Afghanistan / by James Wanliss. : D68-3/322-2014E-PDF

Detrended fluctuation analysis (DFA), a modern technique for examination of complex systems, was applied to combat related data in Afghanistan for the epoch 2002-2009. To detect long-term correlations in the presence of trends, we apply DFA that is able to systematically detect and overcome nonstationarities in the data at all timescales. The objective was to determine whether the nature of combat in Afghanistan, as observed by NATO forces, is fractal in its statistical nature. In every instance we found strong power law correlations in the data, and were able to extract accurate scaling exponents. On the other hand, a decrease in hostilities is likely to persist from one day to the next. We find a measure of predictability inherent in the dynamics of the combat system - there is a history or memory in the signal so that the future dynamics are not random but correlated with past events. This is seen most strongly for Att and d events, and only weakly for is and id events.

Lien permanent pour cette publication :
publications.gc.ca/pub?id=9.821051&sl=1

• Format MARC XML
• Format MARC HTML

Demander des formats alternatifs