غدیر مهدوی

One of the innovations that has been formed in the insurance industry in recent years is transfer the risk to the capital markets. Today, this possibility is provided by issuing insurance bonds and Catastrophe bonds, which are a most important type of insurance-linked securities (ILS), can redress inefficiency in the insurance industry.Today, more and more Catastrophe (CAT) bonds are being issued worldwide, which is welcomed by investors and insurance companies. On the other hand, traditional insurance solutions to cover the risks of Iran's oil and gas industry is not efficient and sufficient and using CAT bonds to transfer risks of this industry to capital markets is a necessary and inevitable issue.The aim of this research is to identify effective factors for issuing Catastrophe bonds in Iran's oil and gas industry. On this basis and after reviewing the literature through library studies, 33 factors were identified in the form of seven categories, based on the similarities. Then, based on Delphi method, experts were asked to express their opinions through an iterative questionnaire. After take the experts' opinions in every round, the statistics analysis was performed and the Delphi process was stopped in the third round. Based on the results, the number of 32 factors in six categories with the titles Legislation and Amendment of the Rules, Knowledge Management, Process Management, Transparency, Creation and Strengthening of Software Platforms and Cultivation were approved by experts and identified as effective factors for issuance of Cat bonds in Iran's oil and gas industry.

This paper investigates insurance pricing using frequency-severity models, focusing on the Potential Deviation Ratio (PDR) as a key measure in premium determination. The objective is to develop an accurate and actuarially sound pricing approach by modeling claim frequency and severity using appropriate statistical distributions. Specifically, we propose that the product of two fitted distributions, one for frequency and the other for severity, can be used to calculate the pure premium. The methodology includes fitting various distributions to frequency and severity data from a major Iranian insurance company. To model the dependency structure, we employ the Clayton copula. This dependency structure is used when calculating pure premium by multiplying frequency and severity samples based on the fitted distributions. Kolmogorov-Smirnov, Anderson-Darling, and Chi-Squared tests are applied to evaluate model performance. The results demonstrate that using the pure premium, based on their respective fitted distributions, significantly enhances the accuracy of the model. This approach leads to more precise risk classification, improved premium setting, and ultimately, fairer pricing strategies. Additionally, the findings indicate that incorporating dependency structures between frequency and severity not only improves risk assessment but also contributes to greater financial stability for insurers.

This study tackles the challenge of customer churn in life insurance, which leads to substantial financial losses. It introduces a transparent, reproducible, and leakage-free machine learning framework designed to identify at-risk policyholders accurately and efficiently. Using 20,000 anonymized Iranian life insurance policies with a churn rate of 26%, the study develops a complete modeling pipeline that includes imputation, standardization, and encoding steps performed strictly within the training process to prevent data leakage. Three model types, logistic regression, random forest, and XGBoost, were trained and evaluated using stratified cross-validation, F1-optimized thresholds, and performance metrics such as ROC-AUC, F1, and precision. Results show that all models perform similarly (ROC-AUC ≈ 0.70), with logistic regression achieving the highest F1 score (0.66) and XGBoost offering the best precision (0.68). The top-ranked predictions captured about 69% of churners, demonstrating strong operational potential. Policyholders with major payment delays were identified as the most churn-prone and easily detectable group. Overall, the findings confirm that transparent and interpretable machine learning models can effectively balance accuracy, simplicity, and practicality, supporting data-driven, value-focused customer retention strategies in the life insurance industry