Asian Journal of Probability and Statistics

On the Logistic and Probit Regression Modelling of Infant Survival at Birth

Favour Chijindu Eke — 2026-05-15

This study focused on modeling the probability of infant survival at birth using the Logistic and Probit regression models. Status of Infant at birth was the response variable; while Systolic BP, Diastolic BP, Age of Mother, Sex of Baby, Weight of Mother, Mode of Delivery, Age/Week Gestation, Parity, and Weight of Baby were the explanatory variables. The Study Data used for illustration comprised of live deliveries at the Federal Government of Nigeria Model Primary Healthcare Center located at Isu-Njaba Town in Isu L.G.A. of Imo State, Nigeria. The test for relationship between any two of the variables used in this study showed that there were negative significant associations between Status of Infant at Birth and Mode of Delivery, and Birth Weight and Mode of Delivery; while there were positive significant associations between Systolic BP and Diastolic BP, Age of Mother and Systolic BP, Gestation and Systolic BP, Gestation and Age of Mother, Diastolic BP and Weight of Mother, Parity and Age of Mother, Parity and Weight of Mother, Parity and Diastolic BP, Birth Weight and Weight of Mother, and Birth Weight and Parity. Multiple logistic and multiple Probit regression models were fitted on the Study Data, and the results of the goodness of fit test using Likelihood Ratio Test at 5% level of significance showed that both models were of good fit. The result of the Wald test showed that Weight of Mother, Mode of Delivery and Birth Weight were significant to the response variable (Status of Infant at Birth – Dead or Alive) at 5% level of significance. The study concluded that Infant Survival at Birth was significantly influenced by Weight of Mother, Mode of Delivery and Birth Weight. Also, both the Logistic and Probit regression models were relatively the same; with AIC values of 164.0844 and 164.2984, respectively.

Analysis of a Two-server Queue with Consultation by Main Server

T. Resmi — 2026-05-15

Multi-server queueing systems with consultation arise in many practical service environments. In such systems, an experienced main server provides assistance to other servers when required. These models are particularly relevant for analyzing the performance of complex and interdependent service operations. This paper investigates a two-server queueing system consisting a main server and a regular server, where the main server provides consultation to the regular server while simultaneously serving its own customers under a preemptive priority discipline. Consequently, the service process at the main server is subject to interruptions triggered by consultation requests. The external arrival process and the consultation request process are modeled as mutually independent Poisson processes. Service times at both the main server and the regular server are assumed to be independent and exponentially distributed. Under these assumptions, the underlying stochastic process is formulated and the stability condition for the system is derived. Key performance measures are obtained and evaluated numerically to illustrate the impact of system parameters.

Three-Warehouse Inventory Model for Non-Instantaneously Deteriorating Items with Ramp-Type Demand, Carbon Cap and Trade, Exponential Bac and Two-Part Trade Credit under Inflation

Ashish Kumar Routray — 2026-05-21

Inventory management is a key supply chain function that has evolved from classical EOQ models to more realistic approaches that consider product deterioration and other practical factors. This paper develops a sustainable three-warehouse inventory model for non-instantaneously deteriorating items under inflationary conditions, incorporating four key modelling innovations relative to the classical two-warehouse benchmark. First, demand is modelled as a ramp-type, time and price-dependent function D(t, p) = (a + bt) p⁻η, capturing simultaneous market growth and iso-elastic price sensitivity. Second, partial shortage backlogging is governed by an exponential time-dependent rate B(t) = e⁻κ(T−t), reflecting increasing customer impatience during prolonged stock-out periods. Third, environmental regulation is modelled via a carbon cap-and-trade (CCT) mechanism with an explicit emission cap and market-based trading price. Fourth, a two-part trade credit arrangement replaces the hybrid instalment scheme: the supplier extends a credit period M to the retailer, who in turn offers customers a credit period N (N ≤ M). Interest earned and charged are derived analytically. Optimal decision variables comprising three depletion times, selling price, and preservation investment are obtained through first-order conditions verified via the Hessian matrix. Numerical experiments and a five-factor sensitivity analysis confirm the model’s robustness and provide actionable managerial insights.

Health Infrastructure Inequality in Sub‑Saharan Africa: A Harmonised and Spatially Validated Facility Approach

Francis Ayiah-Mensah — 2026-05-22

This study addresses critical limitations in large-scale geospatial health facility datasets arising from taxonomy inconsistency and spatial misclassification. The objective was to develop and evaluate an integrated framework that combines taxonomy harmonisation and administrative validation to improve analytical reliability in Sub-Saharan Africa. A multicountry dataset comprising 98,745 facilities was standardised into harmonised ownership and facility-type categories. Spatial validation was performed via country and first administrative (admin1) boundary containment, alongside multimethod outlier detection. Analytical approaches included descriptive statistics, multilevel logistic regression, and spatial clustering (Moran’s I and Getis-Ord Gi*). It appears that taxonomy harmonisation successfully minimised the fragmentation in the classification of facilities with public ownership, representing 57.2%, while 38.1% were not classified. Spatial consistency was high at the country level (≈98%), and was relatively low at the admin1 level (≈23%). Spatial analysis showed that there was a high concentration (Moran's I > 0.5), and facility density was not uniform across countries. The results were seen to be stable across the different pre-processing conditions and produced strong results. The study showed that data harmonisation and spatial validation in analytical workflows enhance the reproducibility and interpretability. The results in this study underscore the importance of scalable, geospatial health data systems to improve evidence-based planning and address health infrastructure disparities.

Transmuted Pareto Exponential Distribution and Its Applications

Md. Shohel Rana — 2026-05-23

In this paper, a new Transmuted Pareto Exponential (TPE “in short”) distribution is illustrated in detail. Parameters are estimated by using MLE methods and a simulation study is conducted for justification of validation of the estimation process. The real life applications of the proposed model is illustrated using two practical life data sets. The empirical implementation shows that the proposed model is the best fitted model than other models used in this study.

On a Dual Risk Model with Dependent Structure and Proportional Gains

Xinran Feng — 2026-05-25

This paper investigates a class of dual risk models featuring a dependent structure and a proportional gain mechanism. It is assumed that a dependency exists between the inter-arrival times of gains and the sizes of the gains, and that the surplus increases proportionally upon each gain event. Using renewal arguments, integral equations for the ruin probability and the ruin time are established. Series expressions for the ruin probability and the Laplace transform of the ruin time are derived by means of the Laplace transform, and the existence and convergence of the iterative solutions are proved. Numerical analysis reveals the effects of the proportional gain coefficient, the arrival rate, and the dependence parameter on the ruin probability. The results indicate that a higher gain frequency and a stronger positive dependence significantly reduce the ruin risk.

Inference for Two Weibull Populations under Joint Ranked Set Sampling Schemes

Mahesh K. Bhingikar — 2026-05-27

In this paper, we develop maximum likelihood (ML) estimation procedures and derive the Fisher information matrix for two Weibull populations under Joint Modified Maximum Ranked Set Sampling (JMxRSS), a sampling framework in which units from two populations are jointly ranked and selected for measurement. The Fisher information matrix is derived to facilitate statistical inference for the ML estimators. We also develop the likelihood ratio test (LRT) for testing the equality of scale parameters of two Weibull populations when the shape parameters are assumed to be known. The performance of the proposed methodology is evaluated under Joint Modified Maximum Ranked Set Sampling (JMxRSS), Joint Modified Minimum Ranked Set Sampling (JMnRSS), and Joint Simple Random Sampling (JSRS). A Monte Carlo simulation study conducted in RStudio demonstrates that the joint ranked set sampling schemes generally outperform JSRS in terms of ML estimation efficiency and statistical power of the LRT. Moreover, the proposed sampling schemes achieve reliable inference with fewer measured observations, highlighting their cost-effectiveness in situations where data collection is expensive or time-consuming. A real-data analysis further supports the practical applicability of the proposed methods and yields conclusions consistent with the simulation results.

Long-Run Price Convergence and Market Efficiency in Selected Selected Cryptocurrencies: A Cointegration and Vector Error Correction Model Approach

Sunday John Oyinloye — 2026-05-28

This study investigates long-run price convergence and weak-form market efficiency among major cryptocurrencies using daily closing prices for Bitcoin (BTC), Ethereum (ETH), Binance Coin (BNB), Solana (SOL), and Ripple (XRP) from April 11, 2020 to August 30, 2025. Unit root tests (ADF, PP, and DF-GLS) indicate that all price series are non-stationary in levels but stationary in first differences, confirming that they are integrated of order one, I(1). Johansen cointegration tests provide evidence of one cointegrating relationship (r ≤ 1) at the 5% significance level (trace statistic = 72.10 > 69.82), indicating a shared long-run equilibrium among the five cryptocurrencies. A Vector Error Correction Model (VECM) with rank one reveals statistically significant error-correction coefficients for all assets, with adjustment speeds of α = 0.005 for BTC, 0.007 for ETH, 0.013 for BNB, 0.017 for SOL, and 0.008 for XRP, implying slow and heterogeneous convergence toward equilibrium. The cointegration spread is highly persistent, with an estimated autoregressive coefficient of ρ = 0.995, corresponding to a half-life of approximately 151 days. Weak-form market efficiency tests based on Ljung-Box statistics and AR(1) regressions indicate that BTC and XRP exhibit return behavior consistent with weak-form efficiency, while ETH, BNB, and SOL display statistically significant short-run predictability. Overall, the results suggest that major cryptocurrency markets are strongly integrated in the long run but exhibit partial weak-form efficiency, with limited long-horizon diversification benefits and slow equilibrium adjustment.

Multi-constrains Logistic Growth Model: An Application to Tourism Carrying Capacity

Raghawendra Mishra — 2026-06-01

This original research article is intended to develop logistic growth models addressing multiple constrains, as J-shape growth have no constrains, S-shape growth model have only one constrain i.e availability of food, but in businesses and economic growth models there are multiple constrains, this article aims to find mathematical model of growth with multiple constrains.

A Dependent Risk Model with Constant Dividend Barrier and Stochastic Premium Income

Nan Huang — 2026-06-01

The paper considers the Gerber-Shiu discounted penalty function for a class of dependent risk models with a constant dividend barrier and stochastic premium income. The system of integro-differential equations satisfied by the Gerber-Shiu discounted penalty function is derived. In particular, an analytical expression for the Laplace transform of ruin time is derived under exponential claim conditions. Finally, numerical examples are presented.

\(\mathcal{L}\)-transformed Tensor Autoregressive Modeling under Heavy-tailed Errors

Nduka, Uchenna Chinedu — 2026-06-03

High dimensional multiple time series data arise in many scientific and financial applications, where complex structures are profound. This paper extends the frequency-domain approach for high-dimensional time series forecasting through data preprocessing using discrete cosine transformation (DCT) combined with heavy-tailed vector autoregressive (VAR) modeling. The proposed framework transforms a tensor of multiple time series using DCT, fits VAR models to the resulting frontal slices, and maps the estimated parameters back to the original domain through inverse DCT. The expectation maximization conditional either (ECME) algorithm provides robust parameter estimation to enable efficient inference under heavy-tailed error distribution. Simulation studies show that more accurate parameter recovery is consistently achieved with ECME than ordinary least squares across different settings, despite recording Frobenius norms of the estimation error that are not close to zero. The practical benefits of the proposed framework are demonstrated through empirical applications using sea surface temperature and prices of 50 randomly selected stocks contained in NASDAQ-100. Results of the analysis show improved forecast performance, with the improvement most pronounced at longer horizons. The gains of integrating frequency-domain transformation with robust estimation procedures, which are crucial for efficient modeling of complex dependencies in multiple time series, are the highlights of the findings of this study.