Abstract
PARAMETER ESTIMATION FROM NEGATIVE-LINDLEY BINOMIAL DISTRIBUTION WITH MAXIMUM POSSIBLE METHODS: APPLICATION IN MOTOR VEHICLE INSURANCE
Yudhistira Naufal Sulaeman, Berlian Setiawaty, I. Gusti Putu Purnaba and Tita Rosita*
ABSTRACT
Frequency modeling of insurance claims plays an important role in an insurance company. This paper explains about negative binomial-Lindley distribution. This distribution has the greatest probability value at zero frequency. Negative binomial-Lindley distribution is usually used to model overdispersed data with large zero frequency probability which is often found in motor vehicle insurance data. For its application, the distribution parameters are estimated using the maximum likelihood method. The numerical calculation has done by using Wolfram Mathematica 11.2 and R 3.5.1 software. The result shows that negative binomial-Lindley distribution is suitable for estimating the frequency of motor vehicle insurance claims with a very good degree of accuracy.
[Full Text Article] [Download Certificate]