Generalized Bernoulli process: simulation, estimation, and application

Abstract

A generalized Bernoulli process (GBP) is a stationary process consisting of binary variables that can capture long-memory property. In this paper, we propose a simulation method for a sample path of GBP and an estimation method for the parameters in GBP. Method of moments estimation and maximum likelihood estimation are compared through empirical results from simulation. Application of GBP in earthquake data during the years of 1800-2020 in the region of conterminous U.S. is provided.

Keywords

• generalized bernoulli process
• bernoulli process
• long-range dependence
• hurst exponent
• earthquake data
• 60g10
• 60g20