Yazar "Erisoglu, Ulku" seçeneğine göre listele

Listeleniyor 1 - 5 / 5
  • Küçük Resim Yok
    Öğe
    A comparison of the parameter estimation methods for bimodal mixture Weibull distribution with complete data
    (Taylor & Francis Ltd, 2015) Karakoca, Aydin; Erisoglu, Ulku; Erisoglu, Murat
    Bimodal mixture Weibull distribution being a special case of mixture Weibull distribution has been used recently as a suitable model for heterogeneous data sets in many practical applications. The bimodal mixture Weibull term represents a mixture of two Weibull distributions. Although many estimation methods have been proposed for the bimodal mixture Weibull distribution, there is not a comprehensive comparison. This paper presents a detailed comparison of five kinds of numerical methods, such as maximum likelihood estimation, least-squares method, method of moments, method of logarithmic moments and percentile method (PM) in terms of several criteria by simulation study. Also parameter estimation methods are applied to real data.
  • Küçük Resim Yok
    Öğe
    Determination of Wind Potential by Two Components Mixture Probability Distribution Models in the Ankara, Turkey
    (Int Journal Renewable Energy Research, 2020) Servi, Tayfun; Gunduz, Selim; Erisoglu, Ulku; Yalcin, Levent
    In this study, hourly average wind speed data in the Ankara, Turkey are modeled with Weibull, Gamma and Rayleigh probability distribution and theirs two component mixture probability distributions. Expectation-Maximization (EM) algorithm is introduced for Maximum Likelihood Estimation (MLE) of mixture probability distributions used in modeling wind speed data. In comparing the modeling performances of probability distributions, the Akaike information criteria, the coefficient of determination, the root of the mean squares and chi-square criteria were used as comparison criteria. Also, in this study, the success in estimation of wind potential was evaluated with relative error. In the study results, it was observed that the mixture distribution models obtained from two different distributions were more successful in modeling wind speed data. The results obtained from the study revealed that the wind potential of Kecioren/Baglum region is higher than Cankaya/Caldag region. According to the wind speed data observed in Kecioren/Baglum and Cankaya/Caldag regions, wind power densities per unit area were calculated as 153.926 W/m(2), and 62.785 W/m(2), respectively.
  • Küçük Resim Yok
    Öğe
    Heterogeneous data modeling with two-component Weibull-Poisson distribution
    (Taylor & Francis Ltd, 2013) Erisoglu, Ulku; Erisoglu, Murat; Calis, Nazif
    The mixture distribution models are more useful than pure distributions in modeling of heterogeneous data sets. The aim of this paper is to propose mixture of Weibull-Poisson (WP) distributions to model heterogeneous data sets for the first time. So, a powerful alternative mixture distribution is created for modeling of the heterogeneous data sets. In the study, many features of the proposed mixture of WP distributions are examined. Also, the expectation maximization (EM) algorithm is used to determine the maximum-likelihood estimates of the parameters, and the simulation study is conducted for evaluating the performance of the proposed EM scheme. Applications for two real heterogeneous data sets are given to show the flexibility and potentiality of the new mixture distribution.
  • Küçük Resim Yok
    Öğe
    Percentile Estimators for Two-Component Mixture Distribution Models
    (Springer International Publishing Ag, 2019) Erisoglu, Ulku; Erisoglu, Murat
    The percentile estimators have a widespread usage in the estimation of distribution parameters because of simplicity and ease of computation. In this study, we investigate the percentile method for two-component mixture distribution models which are commonly used in modeling of heterogeneous univariate data sets. We have proposed percentile estimator for two-component mixture Weibull and two-component mixture Rayleigh distributions according to two different approaches. Performances of the defined percentile estimators were compared with maximum likelihood estimators using simulation. For this purpose, we used several criteria which are bias, mean squared error, mean absolute deviation, mean relative total error and running time of the algorithm. The benefits of the proposed methods have been illustrated by three different real data sets.
  • Küçük Resim Yok
    Öğe
    Value at risk for risk management with normal-logDagum mixture distribution
    (Taylor & Francis Ltd, 2021) Erisoglu, Ulku; Koroglu, Yasemin
    In this article, we aimed to calculate the value at risk (VaR), which is one of the financial risk calculation methods, by using mixture of two different distributions when the financial data does not fit the normal distribution. The normal-logDagum distribution consisting of mixture of Normal and log-Dagum distributions is proposed to calculate the VaR for non-normal financial data in the study. The expected-maximization (EM) algorithm for the maximum likelihood estimates of the parameters of normal-logDagum was defined. In application, the stocks of bank and telecommunication companies were examined. VaR values obtained with different distributions are compared numerically. As a result of the comparison, it was seen that the modeling based on normal-logDagum distribution is more successful in the statistical modeling of financial data.