İsmail Kinaci
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
İsmail Kinaci has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Coşkun Kuş
Selcuk University
5 shared publications - Yunus Akdoğan
Selçuk Üniversitesi, Fen Fakültesi
3 shared publications - Fahreddin Kalkan
Selcuk Universitesi
2 shared publications - Kadir Karakaya
Selçuk Üniversitesi, Fen Fakültesi
2 shared publications Ahmet Pekgör
Faculty of Science
1 shared publication- Buğra SARAÇOĞLU
Selcuk University
1 shared publication
Latest Publications
- 4 October 2026
- Published by Hacettepe University in Hacettepe Journal of Mathematics and Statistics
- No. Advanced O,p. 1-14
Pre-absorption dynamics in multistate run processes: exact distributions and actuarial applications
This paper provides a complete analytical characterization of the joint distribution of state counts before run-based absorption in multistate discrete-time processes. For sequences of independent and identically distributed (i.i.d.) random variables, we derive the exact joint distribution, revealing an elegant geometric–multinomial structure with positive dependence. We then generalize this framework to Markov-dependent processes using matrix-analytic methods, providing a unified approach for both memoryless and memory-based systems. A comparative numerical study, motivated by the classification of lognormal insurance claims, demonstrates the utility of the analysis. It reveals how process memory, such as claim contagion, fundamentally alters pre-absorption statistics and risk profiles compared to the simpler i.i.d. baseline. These results provide a rigorous foundation for developing early-warning signals in dynamic risk management, allowing insurers to design adaptive reinsurance strategies and proactively manage solvency capital.
- 4 October 2026
- Published by Hacettepe University in Hacettepe Journal of Mathematics and Statistics
- No. Advanced O,p. 1-14
Pre-absorption dynamics in multistate run processes: exact distributions and actuarial applications
This paper provides a complete analytical characterization of the joint distribution of state counts before run-based absorption in multistate discrete-time processes. For sequences of independent and identically distributed (i.i.d.) random variables, we derive the exact joint distribution, revealing an elegant geometric–multinomial structure with positive dependence. We then generalize this framework to Markov-dependent processes using matrix-analytic methods, providing a unified approach for both memoryless and memory-based systems. A comparative numerical study, motivated by the classification of lognormal insurance claims, demonstrates the utility of the analysis. It reveals how process memory, such as claim contagion, fundamentally alters pre-absorption statistics and risk profiles compared to the simpler i.i.d. baseline. These results provide a rigorous foundation for developing early-warning signals in dynamic risk management, allowing insurers to design adaptive reinsurance strategies and proactively manage solvency capital.
- 29 April 2026
- Published by Cumhuriyet University in Cumhuriyet Science Journal
- Vol. 47 (2) , 390-394
On the Distribution of Claims for a Process Terminating on a Run of Critical Events
In modern actuarial risk management, the temporal clustering of severe claims is as critical as their cumulative financial magnitude. This study investigates a stochastic risk process that terminates upon the occurrence of a run of consecutive claims exceeding a predefined critical threshold. First, using recursive conditioning techniques, we derive the exact moment generating function for the total severity of exceedances accumulated prior to termination, providing an explicit probability density function for the exponential case when . Second, we determine the exact cumulative distribution and probability mass functions for the maximum number of consecutive non-exceedances observed between two critical claims via first-order linear difference equations. The derived analytical expressions bridge the theory of runs and practical risk management, offering direct tools for dynamic solvency monitoring and operational stress testing without relying on asymptotic approximations
- 29 April 2026
- Published by Cumhuriyet University in Cumhuriyet Science Journal
- Vol. 47 (2) , 390-394
On the Distribution of Claims for a Process Terminating on a Run of Critical Events
In modern actuarial risk management, the temporal clustering of severe claims is as critical as their cumulative financial magnitude. This study investigates a stochastic risk process that terminates upon the occurrence of a run of consecutive claims exceeding a predefined critical threshold. First, using recursive conditioning techniques, we derive the exact moment generating function for the total severity of exceedances accumulated prior to termination, providing an explicit probability density function for the exponential case when . Second, we determine the exact cumulative distribution and probability mass functions for the maximum number of consecutive non-exceedances observed between two critical claims via first-order linear difference equations. The derived analytical expressions bridge the theory of runs and practical risk management, offering direct tools for dynamic solvency monitoring and operational stress testing without relying on asymptotic approximations
- Open Access
- 7 March 2024
- Published by Punjab University Press in Pakistan Journal of Statistics and Operation Research
- Vol. 20 (1) , 37-47
Statistical Inference on Process Capability Index Cpyk for Inverse Rayleigh Distribution under Progressive Censoring
In quality engineering, process capability indexes are used to determine the capability of a process. The well-known of the process capability indexes are Cp, Cpk, Cpm, and Cpmk. These indexes assume the normality of the product lifetime. \citet{maiti2010generalizing} suggested a Cpyk as a generalized process capability index without distributional assumption. In this paper, the maximum likelihood and Bayesian inference on the Cpyk are studied under progressive censoring when the underlying distribution is inverse Rayleigh distribution. Furthermore, Bayesian credible and highest posterior density intervals are discussed with the MCMC procedure. Several types of bootsrap confidence intervals are also considered. A Monte Carlo simulation is conducted in terms of the coverage probabilities and mean lengths of the proposed intervals. An illustrative example is presented to close the paper.
- 7 March 2024
- Published by Punjab University Press in Pakistan Journal of Statistics and Operation Research
- Vol. 20 (1) , 37-47
Open AccessStatistical Inference on Process Capability Index Cpyk for Inverse Rayleigh Distribution under Progressive Censoring
In quality engineering, process capability indexes are used to determine the capability of a process. The well-known of the process capability indexes are Cp, Cpk, Cpm, and Cpmk. These indexes assume the normality of the product lifetime. \citet{maiti2010generalizing} suggested a Cpyk as a generalized process capability index without distributional assumption. In this paper, the maximum likelihood and Bayesian inference on the Cpyk are studied under progressive censoring when the underlying distribution is inverse Rayleigh distribution. Furthermore, Bayesian credible and highest posterior density intervals are discussed with the MCMC procedure. Several types of bootsrap confidence intervals are also considered. A Monte Carlo simulation is conducted in terms of the coverage probabilities and mean lengths of the proposed intervals. An illustrative example is presented to close the paper.
- Open Access
- 1 December 2019
- Published by Gazi University Journal of Science in Gazi University Journal of Science
- Vol. 32 (4) , 1339-1354
Geometric-Zero Truncated Poisson Distribution: Properties and Applications
In this paper, a new discrete distribution is introduced by compounding the geometric distribution with a zero truncated Poisson distribution, named geometric-zero truncated Poisson (GZTP) distribution. Some basic properties of the new distribution, such as the hazard rate function, moments, mode, median, etc., are studied. We show mathematically and numerically that the hazard rate function is increasing. The model parameters are estimated by the moment, least squared error and maximum likelihood methods. A simulation study is performed to compare the performance of the different estimators in terms of bias and mean squared error. An application of the new model is also illustrated using the three real data sets.
- 1 December 2019
- Published by Gazi University Journal of Science in Gazi University Journal of Science
- Vol. 32 (4) , 1339-1354
Open AccessGeometric-Zero Truncated Poisson Distribution: Properties and Applications
In this paper, a new discrete distribution is introduced by compounding the geometric distribution with a zero truncated Poisson distribution, named geometric-zero truncated Poisson (GZTP) distribution. Some basic properties of the new distribution, such as the hazard rate function, moments, mode, median, etc., are studied. We show mathematically and numerically that the hazard rate function is increasing. The model parameters are estimated by the moment, least squared error and maximum likelihood methods. A simulation study is performed to compare the performance of the different estimators in terms of bias and mean squared error. An application of the new model is also illustrated using the three real data sets.
- Open Access
- 30 June 2019
- Published by Cumhuriyet University in Cumhuriyet Science Journal
- Vol. 40 (2) , 378-387
APT-Pareto Distribution and its Properties
Recently, the APT-family has been introduced as a new family of distributions. A special case of this family with exponential distribution is examined in details. In this paper, Pareto is considered as a baseline distribution in APT-family. Several properties of the APT-Pareto distribution such as the moments, hazard rate and survival functions are derived. The maximum likelihood and least square methods are discussed. Simulation study is also performed to get the bias and mean square errors of estimates. A numerical example is given to illustrate the capability of APT-Pareto distribution for modelling real data.
- 30 June 2019
- Published by Cumhuriyet University in Cumhuriyet Science Journal
- Vol. 40 (2) , 378-387
Open AccessAPT-Pareto Distribution and its Properties
Recently, the APT-family has been introduced as a new family of distributions. A special case of this family with exponential distribution is examined in details. In this paper, Pareto is considered as a baseline distribution in APT-family. Several properties of the APT-Pareto distribution such as the moments, hazard rate and survival functions are derived. The maximum likelihood and least square methods are discussed. Simulation study is also performed to get the bias and mean square errors of estimates. A numerical example is given to illustrate the capability of APT-Pareto distribution for modelling real data.
- Open Access
- 30 June 2019
- Published by Cumhuriyet University in Cumhuriyet Science Journal
- Vol. 40 (2) , 493-504
Discriminating between the Lognormal and Weibull Distributions under Progressive Censoring
In this paper, the ratio of maximized likelihood and Minimized Kullback-Leibler Divergence methods are discussed for discrimination between log-normal and Weibull distributions. The progressive Type-II right censored sample is considered in the study. The probability of correct selections is simulated and compared to investigate the performance of the procedures for different censoring schemes and parameter settings.
- 30 June 2019
- Published by Cumhuriyet University in Cumhuriyet Science Journal
- Vol. 40 (2) , 493-504
Open AccessDiscriminating between the Lognormal and Weibull Distributions under Progressive Censoring
In this paper, the ratio of maximized likelihood and Minimized Kullback-Leibler Divergence methods are discussed for discrimination between log-normal and Weibull distributions. The progressive Type-II right censored sample is considered in the study. The probability of correct selections is simulated and compared to investigate the performance of the procedures for different censoring schemes and parameter settings.