Can hidden Markov models be used for inference about operational risk?

Abstract: This thesis aims to investigate the possibility if hidden Markov models (HMM) can be used for inference about operational risk given financial time series data of Auditchanges and Audit prices. The models tested vary in the number of possible states each underlying latent process can take. All models have been implemented usingR-statisticalsoftware along with the depmixS4 package. From the evaluation of the work, it was shown that there was a clear difference between the states, according to the the types of observation they emitted, for the final model. The thesis shows that the biggest factors affecting operational risk were the number of changes of the trades and the time between those changes. It also showed that it was, in large part, the same trader who carried out all the trades as well as changes and only within the internal department. The final conclusion is therefore that HMMs are possible and appropriate to use for inference about operational risk, but that more labeled data are required to express the models predictive performance.