Thanasate Akkanphudit. The topp-leone discrete laplace distribution. Doctoral Degraee(Statistics). Kasetsart University. Office of the University Library. : Kasetsart University, 2020.
Abstract:
A new Topp-Leone generated (TL-G) family of distributions, which is called the Topp-Leone discrete Laplace (TL-DL) distribution, is proposed. It has a shape parameter α > 0 and a scale parameter 0 < p < 1. The TL-DL is an alternative distribution for the discrete data, which is closed to a symmetric distribution when α is close to equal two. Some mathematical properties of the proposed distribution are also derived, such as the moments, order statistics, and quantile function. The unknown parameters that are α and p are estimated by the maximum likelihood estimation (MLE) and the Bayesian approach. Results show that the root mean squared error (RMSE) of MLE and Bayesian method tend to decrease when sample sizes increase, and the RMSE of the Bayesian estimator is smaller than the RMSE of the MLE method, the RMSE of the Bayesian estimator is greater than the RMSE of the MLE method. When the sample sizes are increasing, the RMSE value is decreased. Application studies are presented to show the efficiency of the proposed distribution to the fitting data where it is compared with the discrete Laplace (DL) and discrete normal (DN) distributions. Results show that the TL-DL distribution is the best fit for these data sets.
Kasetsart University. Office of the University Library