BMS MODEL WITH FREQUENCY AND SEVERITY COMPONENTS

• Main Author: HUSAINI (NIM: 20806004), DAVID
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/11963
• Institution: Institut Teknologi Bandung

Car insurance especially third party liability in advance country is a prerequisite for having an automobile. A system is created to fullfill this need, it is called Bonus Malus System (BMS). A policyholder with claim-free year in this period thus in the next period will drop down in class with smaller premium (gain a Bonus). While a policyholder that have claim(s) in this period thus in the next period will move up in class, depending its claim frequency, wuth higher premium (paid a Malus). A BMS is said to be optimal and financially balanced if the total amount of bonuses is equal to the total amount of maluses, and if each policyholder paid premium proportional to the risk carried to the insurance portfolio.<p>In this thesis BMS will be modelled based on frequency component with Poisson-Exponensial model and severity component with Exponensial-Inversa Gamma and Weibull-Gamma models. Those BMS is simulated for 10,000 policyholder through 40 years period, then the averge total premium each period is observed. BMS model based on only frequency component showed a financially balanced feature. While BMS based on frequency and severity components, although its average total premium is close to the starting premium the average total premium tend high in early period then gardually move down along its period. <br />