Hasil Pencarian  ::  Simpan CSV :: Kembali

"Multi-asset risk modeling describes, in a single volume, the latest and most advanced risk modeling techniques for equities, debt, fixed income, futures and derivatives, commodities, and foreign exchange, as well as advanced algorithmic and electronic risk management. Beginning with the fundamentals of risk mathematics and quantitative risk analysis, the book moves on to discuss the laws in standard models that contributed to the 2008 financial crisis and talks about current and future banking regulation. Importantly, it also explores algorithmic trading, which currently receives sparse attention in the literature. By giving coherent recommendations about which statistical models to use for which asset class, this book makes a real contribution to the sciences of portfolio management and risk management."

"Reasuransi berperan sebagai stabilisator industri asuransi dan dapat menjadi alat yang efektif untuk mengurangi risiko bagi perusahaan asuransi. Sebuah transaksi reasuransi adalah perjanjian dari dua atau lebih pihak yang terdiri dari perusahaan asuransi atau disebut juga cedent dan perusahaan reasuransi. Penentuan polis reasuransi yang optimal untuk kedua belah pihak sering menjadi permasalahan. Dikarenakan kepentingan dari perusahaan asuransi dan perusahaan reasuransi bertentangan, seringkali polis reasuransi yang menarik untuk satu pihak dianggap merugikan untuk pihak lainnya. Oleh karena itu, penelitian ini mempertimbangkan kepentingan dari perusahaan asuransi dan perusahaan reasuransi secara sekaligus dengan berfokus pada penentuan polis reasuransi stop-loss yang Pareto-optimal. Bentuk reasuransi yang optimal direpresentasikan oleh ceded loss function yang optimal, yaitu fungsi loss dari perusahaan reasuransi. Ukuran risiko yang akan digunakan adalah Value at Risk (VaR). Ceded loss function yang Pareto-optimal didapat dengan mencari nilai parameter retensi yang optimal, yaitu suatu nilai yang meminimalkan kombinasi linier VaR dari total kerugian perusahaan asuransi dan perusahaan reasuransi pada tingkat kepercayaan yang berbeda,di bawah prinsip ekspektasi premi. Berdasarkan data klaim yang diperoleh dari perusahaan Asuransi Jiwa Reliance Indonesia, didapatkan bentuk reasuransi stop-loss yang Pareto-optimal dalam beberapa skenario.

---

Reinsurance acts as a stabilizer for the insurance industry and can be an effective tool for reducing risk for insurance companies. A reinsurance transaction is an agreement between two or more parties, consisting of an insurance company (also known as the cedent) and a reinsurance company. Determining the optimal reinsurance policy for both parties often presents a challenge. Since the interests of the insurance company and the reinsurance company can be conflicting, a reinsurance policy that is attractive to one party may be detrimental to the other. Therefore, this research considers the interests of both the insurance company and the reinsurance company simultaneously, focusing on determining a Pareto-optimal stop-loss reinsurance policy. The optimal form of reinsurance is represented by the optimal ceded loss function, which is the loss function of the reinsurance company. The risk measure used is Value at Risk (VaR). The Pareto-optimal ceded loss function is obtained by finding the optimal retention parameter, which is a value that minimizes the linear combination of VaR of the total losses for the insurance company and the reinsurance company at different confidence levels, under the premium expectation principle. Based on claim data obtained from Reliance Life Insurance Indonesia, a Pareto-optimal stop-loss reinsurance form is derived in several scenarios."

"This book is focused on the application of modern statistical methods and models to estimate infectious disease parameters, with software guidance, such as R packages, and with data, and features valuable case studies. It is within this context that the Center for Statistics (CenStat, I-Biostat, Hasselt University) and the Centre for the Evaluation of Vaccination and the Centre for Health Economic Research and Modelling Infectious Diseases (CEV, CHERMID, Vaccine and Infectious Disease Institute, University of Antwerp) have collaborated over the past 15 years. This book demonstrates the past and current research activities of these institutes and can be considered to be a milestone in this collaboration."

"

Demi menjaga stabilitas finansial dan mengefektifkan pengelolaan risiko, perusahaan asuransi akan mereasuransikan sebagian klaim yang ada kepada perusahaan reasuransi. Terdapat dua jenis kontrak reasuransi yang biasa digunakan, yaitu reasuransi quota-share dan stop-loss. Pada reasuransi quota-share, klaim dibagi berdasarkan proporsi yang tetap dan premi reasuransi bergantung pada nilai proporsi tersebut, sedangkan pada reasuransi stop-loss, klaim dibagi berdasarkan retensi klaim. Pada skripsi ini kedua reasuransi tersebut dikombinasikan dengan harapan kedua reasuransi tersebut dapat saling menutupi kekurangan yang ada. Setelah dikombinasikan, untuk mendapatkan pertanggungan yang baik bagi perusahaan asuransi, maka perlu dicari nilai proporsi dan retensi yang optimal. Salah satu caranya adalah dengan mengoptimisasi ukuran risiko. Semakin kecil nilai ukuran risiko, maka semakin kecil juga besar kerugian yang akan ditanggung perusahaan asuransi. Ukuran risiko yang digunakan pada skripsi ini adalah Conditional-Tail-Expectation (CTE) yang memiliki relevansi dengan ukuran risiko Value-at-Risk (VaR), yaitu ukuran risiko yang lebih sering digunakan karena penggunaannya yang sederhana, tetapi memiliki kekurangan dalam memberikan informasi terkait dengan kerugian yang sangat besar. Dihitung dengan menggunakan prinsip nilai ekspektasi, premi reasuransi digunakan sebagai kendala pada optimisasi ukuran risiko dengan CTE yang dilakukan untuk masing-masing kombinasi reasuransi, yaitu kombinasi reasuransi stop-loss setelah quota-share dan quota-share setelah stop-loss. Dengan mengoptimisasi CTE, diperoleh bahwa masing-masing kombinasi reasuransi menghasilkan nilai minimal CTE yang sama, sehingga kedua kombinasi reasuransi sama-sama optimal untuk digunakan oleh perusahaan asuransi. Selain itu, didalam menentukan nilai minimal, kondisi yang digunakan pada optimisasi dengan ukuran risiko CTE berbeda dengan VaR.

 

---

To maintain financial stability and to effectively manage the risk, an insurer will partially reinsure the loss to a reinsurance company. Two most commonly used reinsurance contracts are quota-share and stop-loss. In quota-share, the loss will be split based on a fixed proportion and the reinsurance premium depends on the value of the proportion, while in stop-loss the loss will be split depends on on the retention value. In hope that these two types of reinsurance can cover each other weaknesses, this undergraduate thesis combines both quota-share and stop-loss reinsurance. Subsequently, to get a good coverage for the insurer, it is necessary to find the optimal proportion and retention value. One way to accomplish that is using risk measure optimization. The smaller the value of the risk measure, the smaller the loss that borne by the insurer. The risk measure that used in this undergraduate thesis is Conditional-Tail-Expectation (CTE), which has relevance to Value-at-Risk (VaR), the most common used risk measure in practice, but has a weakness in giving information about the value of an extreme loss. Calculated using the expected value principle, the reinsurance premium is used as a constraint in the CTE optimization for each of the reinsurance combinations, which are stop-loss after quota-share and quota-share after stop-loss. By optimizing CTE, it is found that each combination produces the same minimum CTE value, so both reinsurance combinations are optimal to be used by the insurer. Furthermore, in determining the minimum value, the conditions that are used in optimization using CTE are different from VaR

"

"Regulasi dalam penetapan premi menjadi hal yang berperan penting dalam industri asuransi. Pihak regulator perlu mengestimasi ekspektasi klaim agregat untuk menilai tarif premi yang diajukan oleh perusahaan asuransi. Pemodelan distribusi besar klaim adalah salah satu hal yang penting dalam mengestimasi ekspektasi klaim agregat yang akan dijadikan pertimbangan dalam penilaian tarif premi tersebut. Dalam praktiknya, data besar klaim yang digunakan oleh pihak regulator berbentuk data berkelompok yang diperoleh dari industri (perusahaan) asuransi untuk melindungi privasi data nasabah. Akan tetapi, data berkelompok tidak mencakup informasi yang detail terkait data individual sehingga kurang dapat menggambarkan karakteristik yang sebenarnya dari distribusi besar klaim. Oleh karena itu, pada penelitian ini digunakan metode de-grouping untuk memastikan bahwa prediksi dari model distribusi besar klaim yang diperoleh lebih dapat dipercaya. Langkah awal yang dilakukan pada penelitian ini adalah memodelkan data berkelompok yang dimiliki oleh regulator menggunakan beberapa distribusi parametrik. Parameter dari setiap model diestimasi menggunakan metode maximum likelihood. Selanjutnya model terbaik dipilih dengan melakukan uji Kolmogorov-Smirnov dan perbandingan nilai Akaike Information Criterion (AIC). Kemudian prosedur yang sama akan diulang untuk setiap dataset individu yang dibangkitkan dari data grup melalui tiga metode de-grouping yaitu pendekatan Equal Spacing, Uniform Random Sampling, dan Sampling Loss Amount. Adapun kandidat distibusi parametrik yang digunakan adalah distribusi Exponentiated Weibull, Weibull, Exponential, dan Lognormal. Distribusi distribusi ini dipilih karena pada umumnya, data besar klaim memiliki ekor tebal dan skewness positif. Adapun model distribusi terbaik untuk data grup yang digunakan pada penelitian ini adalah distribusi Lognormal.

---

Rate regulation plays an important role in the insurance industry. Regulators need to estimate the expectation of aggregate claims to assess the premium rates proposed by insurance companies. Modeling the distribution of large claims is crucial in estimating the expectation of aggregate claims, which are considered in the premium rate assessment. In practice, the size of loss data used by regulators is in form of grouped data provided by industry of insurance companies to protect privacy of customer’s data. However, grouped data do not include detailed information about individual data and hence cannot describe the actual characteristic of the size of loss distribution as well as individual data. Therefore, this study uses the de-grouping method to ensure that the prediction obtained from the model distribution of size of loss is more reliable. The first step in this study is to model the grouped data held by regulators using several parametric distributions. The parameters of each model are estimated using the maximum likelihood method. The best model is then selected by conducting the Kolmogorov Smirnov test and comparing the Akaike Information Criterion (AIC) values. The same procedure is repeated for each individual dataset generated from the grouped data through three de-grouping methods: Equal Spacing, Uniform Random Sampling, and Sampling Loss Amount. The candidate parametric distributions used are the Exponentiated Weibull, Weibull, Exponential, and Lognormal distributions. These distributions are chosen because size of loss data generally have heavy tails and positive skewness. The best distribution model for the grouped data used in this study is the Lognormal distribution."

"Seiring meningkatnya tren kecelakaan kerja selama periode 2007-2017 serta berjalannya kembali kegiatan usaha secara normal pascapandemi COVID-19, lini usaha asuransi kompensasi pekerja menjadi sangat potensial untuk dikembangkan. Sebagai komponen penting dalam model bisnis asuransi, severitas klaim perlu diprediksi seakurat mungkin karena berpengaruh terhadap penetapan tarif premi bagi tertanggung serta bermanfaat dalam mekanisme pengamatan klaim selama proses penyelesaian klaim. Proses prediksi ini dikategorikan sebagai masalah regresi yang biasanya ditangani oleh model-model pembelajaran mesin untuk data tabular. Namun dalam perkembangan studi pembelajaran mesin, terdapat upaya untuk memanfaatkan model Convolutional Neural Network (CNN) untuk melakukan prediksi terhadap data tabular dengan cara mentransformasikan data tersebut ke dalam representasi gambarnya, salah satunya melalui algoritma Image Generator for Tabular Data (IGTD). Penelitian ini bertujuan untuk menguji akurasi model CNN berbasis algoritma IGTD dalam memprediksi klaim asuransi kompensasi pekerja serta membandingkan performa model tersebut dengan model Multi-Layer Perceptron, Random Forest, serta eXtreme Gradient Boosting. Hasil simulasi dengan metode repeated holdout sebanyak lima iterasi menunjukkan bahwa model CNN dapat memprediksi klaim dengan baik meskipun secara umum belum mampu menyaingi model-model non-CNN secara signifikan.

---

Along with the increasing trend of work accidents during 2007-2017 period as well as the resumption of business activities normally after the COVID-19 pandemic, the workers’ compensation insurance business line has great potential to be developed. As an important component in the insurance business model, the claim severity needs to be predicted as accurate as possible because it affects the determination of premium rates for the insured and is useful in the claim watching mechanism during the claim settlement process. This prediction process is categorized as a regression problem which is usually handled by machine learning models for tabular data. However, in the development of machine learning studies, there are emerging efforts to utilize the Convolutional Neural Network (CNN) model to predict tabular data by transforming the data into its image representation, one of which is through Image Generator for Tabular Data (IGTD) algorithm. This study aims to test the accuracy of the CNN model based on the IGTD algorithm in predicting workers’ compensation insurance claims and to compare the model performance with the Multi-Layer Perceptron, Random Forest, and eXtreme Gradient Boosting models. The simulation result using the repeated holdout method for five iterations shows that the CNN model can well predict the claims, although in general, it has not been able to significantly compete with non-CNN models."

""Preface Financial and actuarial modeling is an ever-changing field with an increased reliance on statistical techniques. This is seen in the changing of competency exams, especially at the upper levels, where topics include more statistical concepts and techniques. In the years since the first edition was published statistical techniques such as reliability measurement, simulation, regression, and Markov chain modeling have become more prominent. This influx in statistics has put an increased pressure on students to secure both strong mathematical and statistical backgrounds and the knowledge of statistical techniques in order to have successful careers. As in the first edition, this text approaches financial and actuarial modeling from a statistical point of view. The goal of this text is twofold. The first is to provide students and practitioners a source for required mathematical and statistical background. The second is to advance the application and theory of statistics in financial and actuarial modeling. This text presents a unified approach to both financial and actuarial modeling through the utilization of general status structures. Future timedependent financial actions are defined in terms of a status structure that may be either deterministic or stochastic. Deterministic status structures lead to classical interest and annuity models, investment pricing models, and aggregate claim models. Stochastic status structures are used to develop financial and actuarial models, such as surplus models, life insurance, and life annuity models. This edition is updated with the addition of nomenclature and notations standard to the actuarial field"--"