MODELLING OF THE EARTHQUAKE CLOSEST DISTANCE BETWEEN SITE TO SITE USING GENERALIZED EXTREME VALUE DISTRIBUTION (GEV), WEIBULL AND LOGNORMAL CASE STUDY : DATA GEMPA PADA WILAYAH TOBELO, MALUKU UTARA

• Main Author: Kainama, Beatrix
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/21401
• Institution: Institut Teknologi Bandung

Earthquake is a event that may cause a substantial losses. That’s why, it is important to estimate the amount of losses so that we can determined the amount of premium that received by a insurance company if there’s claim due to earthquake event. We can determined the earthquake insurance premium by knowing the seismic hazard analysis and estimated losses. In this thesis, we can discuss about the closest distance between site to site at Tobelo City, North Mollucas. It is important to determine the distance in order to determine the effects that may be expreinced by the city during the earthquake event and to facilitate the mitigation process. The distance may be determine by using the coordinates of longitude and latitude using Tobelo city as a point location. Then we can modelling the data using generalized extreme value, weibull and lognormal distribution. The best model can be determined by using the value of Akaike’s Information Criterion (AIC) and Schwarz-Bayesian Information Criterion (BIC). Based on the best probability model, we can estimate the probability of an earthquake on a certain time and distance.With significance level ????=0,05 the goodness of fit tests are carried out using Kolmogorov-Smirnov test. We can say that to modelling the closest distance between site to site (the first, second and third closest distance) using Weibull distribution with ?????=0,8992514 and ?????=1,2368811 (the first closest distance), ?????=1,080393 and ?????=0,3838593 (the seond closest distance) and ?????=1,257232 and ?????=3,536596 (the third closest distance).