Shuhrah Alghamdi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Shuhrah Alghamdi has a total of 37 co-authors. Here are the top co-authors by number of shared publications:
- Hassan S. Bakouch
Department of Mathematics, College of Science, Qassim University, Buraydah 51452, Saudi Arabia
6 shared publications - Qasim Ramzan
Department of Statistics University of Sargodha Sargodha Pakistan
3 shared publications - Randa Alharbi
Department of Statistics, Faculty of Science, University of Tabuk, Tabuk 47512, Saudi Arabia
3 shared publications - Ahmed Elshahhat
Faculty of Technology and Development, Zagazig University, Zagazig 44519, Egypt
2 shared publications - Ayman A. Amin
Department of Mathematics, College of Arts and Sciences, Najran University, Najran 66462, Saudi Arabia
2 shared publications - Fernando Antonio Moala
Department of Statistics, Faculty of Science and Technology, Sao Paulo State University, Sao Paulo 19060-900, Brazil
2 shared publications
Latest Publications
- Open Access
- 13 November 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (22) , 4175
Affine Power-Hazard Distributions on a Finite Support: Construction, Shape Theory, and Likelihood Inference
Finite-supportfirst-success models arise when an event must occur by a prescribed terminal trial. The Alê distribution uses the linear hazard , which does not permit independent adjustment of the baseline success propensity and hazard curvature for fixed n. We introduce the affine power-hazard (APH) distribution on , defined by , where , , and . The Alê distribution is recovered at . We derive the main distributional functions, a stochastic first-success representation, stochastic orderings, and limiting properties. The hazard is increasing and is concave, linear, or convex according to , , or . The probability mass function is unimodal for , whereas bimodality may occur for . For , the continuous parameters are structurally identifiable, although identification of becomes weak in the large-power regime, where the model approaches a right-censored geometric distribution. We develop likelihood-based inference for known and unknown endpoints and examine finite-sample estimation and sensitivity to endpoint-search bounds. Two applications illustrate both the usefulness of an externally specified finite endpoint and the inferential limitations arising near the geometric limiting regime.- 13 November 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (22) , 4175
Open AccessAffine Power-Hazard Distributions on a Finite Support: Construction, Shape Theory, and Likelihood Inference
Finite-supportfirst-success models arise when an event must occur by a prescribed terminal trial. The Alê distribution uses the linear hazard , which does not permit independent adjustment of the baseline success propensity and hazard curvature for fixed n. We introduce the affine power-hazard (APH) distribution on , defined by , where , , and . The Alê distribution is recovered at . We derive the main distributional functions, a stochastic first-success representation, stochastic orderings, and limiting properties. The hazard is increasing and is concave, linear, or convex according to , , or . The probability mass function is unimodal for , whereas bimodality may occur for . For , the continuous parameters are structurally identifiable, although identification of becomes weak in the large-power regime, where the model approaches a right-censored geometric distribution. We develop likelihood-based inference for known and unknown endpoints and examine finite-sample estimation and sensitivity to endpoint-search bounds. Two applications illustrate both the usefulness of an externally specified finite endpoint and the inferential limitations arising near the geometric limiting regime. - 7 October 2026
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- p. 1-32
Reliability-oriented CUSUM monitoring of breast cancer features using Burr–X distribution and progressive censoring
1. Accurate and timely monitoring of disease biomarkers is a cornerstone of modern oncology, with significant implications for prognosis and treatment. In breast cancer (BC), continuous biomarkers ...
- 7 October 2026
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- p. 1-32
Reliability-oriented CUSUM monitoring of breast cancer features using Burr–X distribution and progressive censoring
1. Accurate and timely monitoring of disease biomarkers is a cornerstone of modern oncology, with significant implications for prognosis and treatment. In breast cancer (BC), continuous biomarkers ...
- Open Access
Proportional Autoregressive Quasi-Lindley Half-Logistic Unit Process with Application in Modeling Crime Dynamics
The manuscript proposes a novel bounded proportional autoregressive (PAR) process based on the quasi-Lindley half-logistic unit (QHU) distribution, termed the QHU-PAR(1) process. Fundamental probabilistic properties of the process are established, including its Markov structure, moments, and stationarity conditions. A pseudo-innovation-based parameter estimation procedure is performed, demonstrating strong consistency and asymptotic normality of the resulting estimators. A Monte Carlo study is also conducted, showing satisfactory performance of the proposed estimators on finite samples, while the practical utility of the model is illustrated through the analysis of normalized crime-related time series. Comparative results show that the proposed QHU-PAR(1) process outperforms some competing specifications, highlighting its flexibility and efficiency for modeling bounded and asymmetric stochastic phenomena.
Open AccessProportional Autoregressive Quasi-Lindley Half-Logistic Unit Process with Application in Modeling Crime Dynamics
The manuscript proposes a novel bounded proportional autoregressive (PAR) process based on the quasi-Lindley half-logistic unit (QHU) distribution, termed the QHU-PAR(1) process. Fundamental probabilistic properties of the process are established, including its Markov structure, moments, and stationarity conditions. A pseudo-innovation-based parameter estimation procedure is performed, demonstrating strong consistency and asymptotic normality of the resulting estimators. A Monte Carlo study is also conducted, showing satisfactory performance of the proposed estimators on finite samples, while the practical utility of the model is illustrated through the analysis of normalized crime-related time series. Comparative results show that the proposed QHU-PAR(1) process outperforms some competing specifications, highlighting its flexibility and efficiency for modeling bounded and asymmetric stochastic phenomena.
- Open Access
- 2 September 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (17) , 3163
Full Bayesian Analysis of ARX Models Under Scale-Mixtures of Normal Errors: An Application to Solar Radiation in Najran, Saudi Arabia
Autoregressive models with exogenous variables (ARX) constitute a fundamental family of time series tools with broad applicability across engineering, environmental science, and finance. A persistent limitation of standard Bayesian treatments is the Gaussian error assumption, which frequently proves inadequate when dealing with real data displaying heavy tails or occasional extreme observations. To overcome this shortcoming, the present paper develops a complete Bayesian inferential framework for ARX models under the scale-mixtures of normal (SMN) error distribution, integrating model identification, parameter estimation, and multi-step-ahead prediction within a unified scheme. A stochastic search variable selection (SSVS) procedure is adapted to perform simultaneous selection of the autoregressive order and the active exogenous regressors by assigning binary latent indicators to each candidate coefficient. Mixture-of-normals priors are specified for the dynamic and exogenous coefficients and an inverse-gamma prior for the error scale, while Bernoulli priors govern the latent selection indicators. These choices yield tractable full conditional posterior distributions: multivariate normal for the complete coefficient vector, inverse-gamma for the scale, and Bernoulli for the indicators. The conditional predictive distribution of future observations is also multivariate normal. For SMN-specific mixing parameters whose conditionals lack standard forms, Metropolis–Hastings steps are embedded within the Gibbs sampler. An extensive simulation study evaluates recovery accuracy across three SMN distributions and several ARX configurations. The methodology is then applied to the forecasting of daily global horizontal irradiance (GHI) in Najran, southwestern Saudi Arabia, using clear-sky GHI as an exogenous covariate, demonstrating the practical value of the proposed framework in a renewable energy context.
- 2 September 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (17) , 3163
Open AccessFull Bayesian Analysis of ARX Models Under Scale-Mixtures of Normal Errors: An Application to Solar Radiation in Najran, Saudi Arabia
Autoregressive models with exogenous variables (ARX) constitute a fundamental family of time series tools with broad applicability across engineering, environmental science, and finance. A persistent limitation of standard Bayesian treatments is the Gaussian error assumption, which frequently proves inadequate when dealing with real data displaying heavy tails or occasional extreme observations. To overcome this shortcoming, the present paper develops a complete Bayesian inferential framework for ARX models under the scale-mixtures of normal (SMN) error distribution, integrating model identification, parameter estimation, and multi-step-ahead prediction within a unified scheme. A stochastic search variable selection (SSVS) procedure is adapted to perform simultaneous selection of the autoregressive order and the active exogenous regressors by assigning binary latent indicators to each candidate coefficient. Mixture-of-normals priors are specified for the dynamic and exogenous coefficients and an inverse-gamma prior for the error scale, while Bernoulli priors govern the latent selection indicators. These choices yield tractable full conditional posterior distributions: multivariate normal for the complete coefficient vector, inverse-gamma for the scale, and Bernoulli for the indicators. The conditional predictive distribution of future observations is also multivariate normal. For SMN-specific mixing parameters whose conditionals lack standard forms, Metropolis–Hastings steps are embedded within the Gibbs sampler. An extensive simulation study evaluates recovery accuracy across three SMN distributions and several ARX configurations. The methodology is then applied to the forecasting of daily global horizontal irradiance (GHI) in Najran, southwestern Saudi Arabia, using clear-sky GHI as an exogenous covariate, demonstrating the practical value of the proposed framework in a renewable energy context.
- Open Access
- 19 July 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (14) , 2626
Adaptive Type-II Competing Risks Analysis for the Generalized Inverted Exponential Model Using Medical Applications
This study investigates parameter estimation utilizing an adaptive progressive Type-II competing risks framework. In this work, we provide a comprehensive analysis of the statistical features and estimation methods for the Generalized Inverted Exponential (GIE) model. Assuming a population subject to two independent failure causes, both following a GIE distribution, a competing risks model is formulated. Point and interval estimates are derived using both maximum likelihood, with associated asymptotic confidence intervals, and Bayesian approaches via Markov Chain Monte Carlo simulation to obtain credible intervals. The effectiveness of these methods is subsequently demonstrated by applying them to real datasets and through a comprehensive Monte Carlo simulation study designed to assess estimator performance.
- 19 July 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (14) , 2626
Open AccessAdaptive Type-II Competing Risks Analysis for the Generalized Inverted Exponential Model Using Medical Applications
This study investigates parameter estimation utilizing an adaptive progressive Type-II competing risks framework. In this work, we provide a comprehensive analysis of the statistical features and estimation methods for the Generalized Inverted Exponential (GIE) model. Assuming a population subject to two independent failure causes, both following a GIE distribution, a competing risks model is formulated. Point and interval estimates are derived using both maximum likelihood, with associated asymptotic confidence intervals, and Bayesian approaches via Markov Chain Monte Carlo simulation to obtain credible intervals. The effectiveness of these methods is subsequently demonstrated by applying them to real datasets and through a comprehensive Monte Carlo simulation study designed to assess estimator performance.
- Open Access
- 18 June 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (12) , 2188
Bayesian Estimation of Autoregressive Models with Exogenous Variables Under Scale-Mixtures of Normal Errors
Autoregressive models with exogenous variables (ARX) constitute a fundamental class of dynamic regression models used extensively for time series analysis across a wide range of applications. A pervasive limitation of the existing Bayesian analyses of ARX models is their near-exclusive reliance on the Gaussian error assumption, which is routinely violated in empirical applications exhibiting heavy-tailed innovations, distributional outliers, or excess kurtosis. To address this deficiency, we develop a rigorous Bayesian estimation framework for these models whose errors are drawn from the scale-mixtures of normal (SMN) family, which is a rich, symmetric, heavy-tailed class of distributions. Exploiting the hierarchical stochastic representation of the SMN family through observation-specific latent scale-mixing variables, the ARX model is embedded in an augmented data structure that restores Gaussian conditional structure. Under three distinct prior formulations—namely, normal-gamma, Zellner’s g-prior, and Jeffreys’ prior—we derive closed-form full conditional posterior distributions for the ARX coefficient vector and the error scale parameter, which follow multivariate normal and inverse-gamma distributions, respectively. In addition, for the SMN-specific shape parameters, we derive the full conditional posteriors for each distribution in the family, and some of them are non-standard distributions handled by embedding Metropolis-Hastings steps within the Gibbs sampler. The resulting hybrid MCMC algorithm is validated through a comprehensive simulation study spanning three ARX model configurations and all three SMN special cases. A real macroeconomic application to US consumer price inflation demonstrates the practical utility of the framework, confirming heavy-tailed residuals and yielding precise, well-calibrated posterior estimates.
- 18 June 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (12) , 2188
Open AccessBayesian Estimation of Autoregressive Models with Exogenous Variables Under Scale-Mixtures of Normal Errors
Autoregressive models with exogenous variables (ARX) constitute a fundamental class of dynamic regression models used extensively for time series analysis across a wide range of applications. A pervasive limitation of the existing Bayesian analyses of ARX models is their near-exclusive reliance on the Gaussian error assumption, which is routinely violated in empirical applications exhibiting heavy-tailed innovations, distributional outliers, or excess kurtosis. To address this deficiency, we develop a rigorous Bayesian estimation framework for these models whose errors are drawn from the scale-mixtures of normal (SMN) family, which is a rich, symmetric, heavy-tailed class of distributions. Exploiting the hierarchical stochastic representation of the SMN family through observation-specific latent scale-mixing variables, the ARX model is embedded in an augmented data structure that restores Gaussian conditional structure. Under three distinct prior formulations—namely, normal-gamma, Zellner’s g-prior, and Jeffreys’ prior—we derive closed-form full conditional posterior distributions for the ARX coefficient vector and the error scale parameter, which follow multivariate normal and inverse-gamma distributions, respectively. In addition, for the SMN-specific shape parameters, we derive the full conditional posteriors for each distribution in the family, and some of them are non-standard distributions handled by embedding Metropolis-Hastings steps within the Gibbs sampler. The resulting hybrid MCMC algorithm is validated through a comprehensive simulation study spanning three ARX model configurations and all three SMN special cases. A real macroeconomic application to US consumer price inflation demonstrates the practical utility of the framework, confirming heavy-tailed residuals and yielding precise, well-calibrated posterior estimates.
- Open Access
- 4 June 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (11) , 1993
A New Generalized ZLindley Model: Theory, Inference, and Engineering Reliability Applications
This study presents a new version of the ZLindly (ZL) model that improves modeling flexibility while maintaining ease of analysis, allowing for the simultaneous accommodation of redundant zeros, thick-tailed behavior, and complex failure rate dynamics within a unified probabilistic framework. Marshall–Olkin (MO) theory facilitates this advancement. The MOZL hazard rate can exhibit several patterns, including increasing, decreasing, bathtub, or upside-down bathtub-shaped. These features enable the model to capture diverse reliability phenomena such as early-life failures, random shocks, and wear-out effects. Comprehensive theoretical investigations were conducted and shown to be governed by an interpretable dual-parameter mechanism, where the Marshall–Olkin parameter controls tail behavior and dispersion, while the scale parameter regulates skewness and hazard evolution. A likelihood-based approach was developed under Type-II censoring conditions, and rigorous evidence is provided for the existence and uniqueness. To address inferential uncertainty, both classical asymptotic confidence intervals and log-normal approximations were constructed. Within a Bayesian framework, independent gamma priors were assumed, and posterior inference was performed via an efficient Metropolis–Hastings algorithm. Bayesian point and credible estimators were obtained and compared with their classical counterparts. An extensive simulation study demonstrates that Bayesian estimators, particularly with informative priors, consistently outperform likelihood-based estimators in terms of bias, mean squared error, interval length, and coverage probability, especially for moderate sample sizes and higher censoring levels. Three engineering applications are provided to assess the practical utility of the MOZL model, where it provides superior goodness-of-fit relative to 15 competing models, including MO–Exponential, MO–Gompertz, MO–Nadarajah–Haghighi, MO–Exponentiated Weibull, and Birnbaum–Saunders, among others. Overall, the proposed MOZL distribution emerges as a flexible, interpretable, and computationally efficient lifetime model whose structurally meaningful parameter interactions enhance distributional balance and flexible hazard behavior, thereby contributing to modern symmetry-oriented distribution theory while offering valuable applications in reliability engineering, survival analysis, and applied statistical modeling.
- 4 June 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (11) , 1993
Open AccessA New Generalized ZLindley Model: Theory, Inference, and Engineering Reliability Applications
This study presents a new version of the ZLindly (ZL) model that improves modeling flexibility while maintaining ease of analysis, allowing for the simultaneous accommodation of redundant zeros, thick-tailed behavior, and complex failure rate dynamics within a unified probabilistic framework. Marshall–Olkin (MO) theory facilitates this advancement. The MOZL hazard rate can exhibit several patterns, including increasing, decreasing, bathtub, or upside-down bathtub-shaped. These features enable the model to capture diverse reliability phenomena such as early-life failures, random shocks, and wear-out effects. Comprehensive theoretical investigations were conducted and shown to be governed by an interpretable dual-parameter mechanism, where the Marshall–Olkin parameter controls tail behavior and dispersion, while the scale parameter regulates skewness and hazard evolution. A likelihood-based approach was developed under Type-II censoring conditions, and rigorous evidence is provided for the existence and uniqueness. To address inferential uncertainty, both classical asymptotic confidence intervals and log-normal approximations were constructed. Within a Bayesian framework, independent gamma priors were assumed, and posterior inference was performed via an efficient Metropolis–Hastings algorithm. Bayesian point and credible estimators were obtained and compared with their classical counterparts. An extensive simulation study demonstrates that Bayesian estimators, particularly with informative priors, consistently outperform likelihood-based estimators in terms of bias, mean squared error, interval length, and coverage probability, especially for moderate sample sizes and higher censoring levels. Three engineering applications are provided to assess the practical utility of the MOZL model, where it provides superior goodness-of-fit relative to 15 competing models, including MO–Exponential, MO–Gompertz, MO–Nadarajah–Haghighi, MO–Exponentiated Weibull, and Birnbaum–Saunders, among others. Overall, the proposed MOZL distribution emerges as a flexible, interpretable, and computationally efficient lifetime model whose structurally meaningful parameter interactions enhance distributional balance and flexible hazard behavior, thereby contributing to modern symmetry-oriented distribution theory while offering valuable applications in reliability engineering, survival analysis, and applied statistical modeling.
- Open Access
- 23 May 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (11) , 1803
New Power Reliability Modeling via Randomized Progressive First-Failure Beta–Binomial Censoring: Theory, Optimization, and Engineering Applications to Fiber Strengths
In modern reliability engineering, modeling bounded lifetime data under realistic experimental conditions is still challenging, especially when censoring schemes and unit removals are random. This study proposes a new and unified reliability framework by combining the flexible powering new power (PNP) distribution with a grouping-based progressive first-failure mechanism using a beta-binomial random design. The proposed approach explicitly accounts for the randomness in group removals, providing a more realistic description of practical life-testing experiments. Classical estimation is carried out using maximum likelihood methods with the Newton-Raphson algorithm, along with confidence intervals constructed under both standard and log-transformed parameterizations. To increase flexibility in inference, a Bayesian approach is developed based on a joint gamma and shifted log-normal prior, which respects parameter constraints and incorporates prior uncertainty. Since the posterior distributions cannot be obtained in closed form, a Metropolis-Hastings Markov chain Monte Carlo algorithm is used to generate reliable posterior estimates and credible intervals. Additionally, beyond sensitivity analysis, multiple prior robustness diagnostics are incorporated to ensure reliable hyperparameter calibration and to safeguard against prior misspecification. The performance of the proposed estimators is carefully examined through extensive Monte Carlo simulations under different censoring schemes and parameter settings. The simulation results indicate that the proposed Bayesian procedures often provide more stable estimation and shorter interval estimates with competitive coverage probabilities compared with the corresponding classical methods, particularly under moderate-to-heavy censoring settings. To demonstrate its practical usefulness, the proposed model is applied to two real datasets on tensile strength of carbon and polyester fibers, where it provides a good fit and useful insights into material reliability and failure behavior. In the same applications, the practical relevance and superior performance of the proposed distribution are demonstrated, where it outperforms existing bounded versions of several well-known models, including the gamma, Weibull, and Birnbaum-Saunders distributions. Overall, this work contributes to reliability analysis by offering a flexible and computationally efficient framework that accounts for both random censoring and complex lifetime patterns, with potential applications in engineering, materials science, and applied reliability studies.
- 23 May 2026
- Published by MDPI AG in Mathematics
- Vol. 14 (11) , 1803
Open AccessNew Power Reliability Modeling via Randomized Progressive First-Failure Beta–Binomial Censoring: Theory, Optimization, and Engineering Applications to Fiber Strengths
In modern reliability engineering, modeling bounded lifetime data under realistic experimental conditions is still challenging, especially when censoring schemes and unit removals are random. This study proposes a new and unified reliability framework by combining the flexible powering new power (PNP) distribution with a grouping-based progressive first-failure mechanism using a beta-binomial random design. The proposed approach explicitly accounts for the randomness in group removals, providing a more realistic description of practical life-testing experiments. Classical estimation is carried out using maximum likelihood methods with the Newton-Raphson algorithm, along with confidence intervals constructed under both standard and log-transformed parameterizations. To increase flexibility in inference, a Bayesian approach is developed based on a joint gamma and shifted log-normal prior, which respects parameter constraints and incorporates prior uncertainty. Since the posterior distributions cannot be obtained in closed form, a Metropolis-Hastings Markov chain Monte Carlo algorithm is used to generate reliable posterior estimates and credible intervals. Additionally, beyond sensitivity analysis, multiple prior robustness diagnostics are incorporated to ensure reliable hyperparameter calibration and to safeguard against prior misspecification. The performance of the proposed estimators is carefully examined through extensive Monte Carlo simulations under different censoring schemes and parameter settings. The simulation results indicate that the proposed Bayesian procedures often provide more stable estimation and shorter interval estimates with competitive coverage probabilities compared with the corresponding classical methods, particularly under moderate-to-heavy censoring settings. To demonstrate its practical usefulness, the proposed model is applied to two real datasets on tensile strength of carbon and polyester fibers, where it provides a good fit and useful insights into material reliability and failure behavior. In the same applications, the practical relevance and superior performance of the proposed distribution are demonstrated, where it outperforms existing bounded versions of several well-known models, including the gamma, Weibull, and Birnbaum-Saunders distributions. Overall, this work contributes to reliability analysis by offering a flexible and computationally efficient framework that accounts for both random censoring and complex lifetime patterns, with potential applications in engineering, materials science, and applied reliability studies.
- Open Access
Minification Integer-Valued Split-BREAK Process with Power Series Innovations and Application in Fire Safety Dynamics
This manuscript introduces a new class of count time series models, referred to as the minification integer-valued Split-BREAK (MIN–SB) process. The proposed framework extends the Split-BREAK modeling philosophy to the integer-valued setting and provides a flexible mechanism for capturing rare events, zero inflation, and structural regime changes frequently observed in safety-related data. The main stochastic properties of the MIN–SB process are derived, including stationarity conditions, explicit moment structure, and correlation dynamics. A key theoretical result reveals an implicit hidden Markov structure underlying the observable process, providing a structural explanation for zero clustering observed in rare-event count processes. Parameter estimation is developed using a simulated method of moments (SMM) approach based on zero-related statistics, and the asymptotic properties of the resulting estimators are established. A Monte Carlo simulation study demonstrates favorable finite-sample performance of the proposed estimation procedure. The practical usefulness of the model is illustrated through an empirical application to time series of injuries and fatalities caused by fire accidents in Serbia. The results show that the MIN–SB specification provides a flexible and accurate framework for modeling zero-inflated count processes arising in fire safety dynamics.
Open AccessMinification Integer-Valued Split-BREAK Process with Power Series Innovations and Application in Fire Safety Dynamics
This manuscript introduces a new class of count time series models, referred to as the minification integer-valued Split-BREAK (MIN–SB) process. The proposed framework extends the Split-BREAK modeling philosophy to the integer-valued setting and provides a flexible mechanism for capturing rare events, zero inflation, and structural regime changes frequently observed in safety-related data. The main stochastic properties of the MIN–SB process are derived, including stationarity conditions, explicit moment structure, and correlation dynamics. A key theoretical result reveals an implicit hidden Markov structure underlying the observable process, providing a structural explanation for zero clustering observed in rare-event count processes. Parameter estimation is developed using a simulated method of moments (SMM) approach based on zero-related statistics, and the asymptotic properties of the resulting estimators are established. A Monte Carlo simulation study demonstrates favorable finite-sample performance of the proposed estimation procedure. The practical usefulness of the model is illustrated through an empirical application to time series of injuries and fatalities caused by fire accidents in Serbia. The results show that the MIN–SB specification provides a flexible and accurate framework for modeling zero-inflated count processes arising in fire safety dynamics.