Periodic account activity and automated money laundering detection

Abstract

Introduces the use of account activity relative to client peer groups as a means of identifying unusual behaviour, as part of money laundering detection; assumptions can be made about what constitutes normal transactional behaviour for an individual, so that deviations from this can generate risk factors. Analyses an account history as a time series of asset movements that can be characterised by a function whose Fourier transform can be computed, yielding a function of a frequency; the so‐called Power Spectrum can be derived, and this specifies the “power” or magnitude of asset movement in the account as a function of frequency. Points out that the method also allows detailed inter‐institutional comparisons of account activity.

Citation

Young, C. (2004), "Periodic account activity and automated money laundering detection", Journal of Money Laundering Control, Vol. 7 No. 4, pp. 295-297. https://doi.org/10.1108/13685200410810001