Earthquake probability prediction in Sumatra Island using Poisson Hidden Markov Model (HMM)

Abstract

In this paper, the probability of an earthquake occurring in Sumatra Island will be estimated. From the earthquake data that is obtained from USGS for the period of 1 December 2008 until 1 December 2018 in Sumatera Island, it is found that the observed variance is larger than the observed mean, which is called overdispersion. By checking the Autocorrelation Function (ACF) of earthquake data in Sumatera, there is a strong indication that earthquake data in Sumatera is serially-dependent. For these reasons, the Poisson Hidden Markov Model (HMM) is a good model to estimate the probability of earthquake in Sumatra.After the starting values for parameters have been determined, the next step is inputting the starting values to the Expectation-Maximization Algorithm to yield the best model. Then, after getting the best model of earthquake events, the model obtained is used to estimate the probability of earthquake events occurred. The result of this study can be used as one of the steps in disaster mitigation for the people in Sumatra