Early warning model with machine learning for Turkish insurance sector

Koçer, Günay Burak
Early warning models are needed to ensure that relevant stakeholders would not cause worse outcomes by ignoring the risks. Specific to the insurance sector, this risk is about meeting the obligations and its sustainability. In this study, an early warning model is formed using the ratios obtained from the financial statements of insurance companies. The goal of the model is to identify the risk areas and to support the strengthening of the financial structure and taking timely measures in companies. Since the data consisted of annual periods, only non-life insurance companies are included in the analysis. The annual balance sheet and statements of income declared by the Insurance Association of Turkey (IAT) and the annual reports about insurance and private pension activities published by The Republic of Turkey Ministry of Treasury and Finance (TRMTF) are examined based on companies, and ratios to be used in the model are calculated. The data set is composed of 70 financial ratios obtained from the financial statements of all non-life insurance companies operating between 2011-2018. Classifications are determined as credit, liquidity, market, reinsurance, underwriting, technical provisions, reputational, operational, profitability, and capital risks. In the developed model, the values realized in 2018 are estimated with machine learning methods by using the data of 2011-2017. Random Forest, Neural Networks, Gradient Boosting Machine and Extreme Gradient Boosting are used as analysis methods, and Boruta is used as a feature selection method. Capital requirement ratio is chosen as the dependent variable. The other 69 ratios are the independent variables and this set is reduced to 22 independent variables by the Boruta method. Analyzes are conducted on two datasets with 69 and 22 independent variables and the results are compared. Furthermore, after predicting the 2018 values for above mentioned 38 non-life insurance companies and doing stage classifications with those values, whether the actual stages match with the predicted stages is evaluated. The best estimate accuracy belongs to Random Forest and Gradient Boosting Machine methods with 87%. The predictive power decreases with 22 independent variables, but the results are still close to each other. Then, performances of machine learning models are compared over capital adequacy classification. In this comparison, the best predictive method is Neural Networks with 95% accuracy.


Assessment of risk paths in construction projects
Ozcan, G; Dikmen Toker, İrem; Birgönül, Mustafa Talat (2011-01-01)
Risk identification is the most important step in project risk management (PRM), as the risk models are constructed based on previously defined risk sources and their interrelations. Although previous studies have concentrated on a one-way relation between risk factors, there are cause-effect relationships between risk sources, events and consequences. 'Risk paths' represent the causal relationships between risk sources, events and consequences and they should be assessed during risk identification, rather ...
Early versus delayed order commitments in multi-retailer inventory systems
Bakal, İsmail Serdar; Geunes, Joseph (2007-12-01)
We consider the implications of order commitment policies in terms of expected profits under demand uncertainty. Specifically, we investigate a system with multiple retailers and a single supplier in a single period context where one of the retailers is the primary customer of the supplier. In the base model, the primary customer orders after observing demand, and the supplier rations its production to retailers. In the early commitment model, the primary retailer orders before the selling season and receiv...
Stochastic modeling of accident risks associated with an underground coal mine in Turkey
Sari, Mehmet; Kestel, Sevtap Ayşe; Karpuz, Celal; Duzgun, H. Sebnem B. (2009-01-01)
In this study, a methodology is proposed towards development of all uncertainty model that includes randomness in the occurrence of days-lost accidents in a coal mine. The accident/injury data consists of 1390 days-lost accident cases recorded at GLI-Tuncbilek underground lignite mine from January 1994 to December 2002. In the first step of proposed methodology, the frequency and the severity of the accidents have been modeled statistically by fitting appropriate distributions. The test done by BestFit soft...
Identification of Risk Paths in International Construction Projects Using Structural Equation Modeling
Eybpoosh, Matineh; Dikmen Toker, İrem; Birgönül, Mustafa Talat (American Society of Civil Engineers (ASCE), 2011-12-01)
The major aim of this research is to demonstrate that causal relationships exist among various risk factors that necessitate identification of risk paths, rather than individual risk factors, during risk assessment of construction projects. International construction projects have more complex risk-emergence patterns because they are affected by global and foreign country conditions and project-related factors. Identification of a network of interactive risk paths, each of which initiated from diverse vulne...
Performance analysis of open-source DEMs in tsunami inundation modelling
Tüfekçi Enginar, Duygu; Dogan, Gozde Guney; Süzen, Mehmet Lütfi; Yalçıner, Ahmet Cevdet (2022-08-01)
In natural hazard modeling, underestimation is not acceptable, since it will produce unreliable and inadequate information for decision makers to work on during hazard mitigation planning. Including tsunami inundation modeling, result of the natural hazard models is highly sensitive to the accuracy of input data. High resolution elevation data, such as light detection and ranging (LiDAR) Digital Elevation Models (DEMs) can provide modeling results that are closer to the reality. However, gathering such high...
Citation Formats
G. B. Koçer, “Early warning model with machine learning for Turkish insurance sector,” Thesis (M.S.) -- Graduate School of Applied Mathematics. Financial Mathematics., Middle East Technical University, 2019.