UTILITY FUNCTION-BASED LOSS PORTFOLIO: MODELING, VALUATION, AND SELECTION

• Main Author: Sholiha, Aminatus
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/79901
• Institution: Institut Teknologi Bandung

The portfolio is a model that can be used to construct an aggregate of two large losses that are independent and identically distributed. The portfolio model in this study uses gamma and Weibull distributions. Currently, the portfolio model does not consider the preferences of investors. Therefore, there is a need for a function that can accommodate the flexibility of investor preferences. In particular, the function used is a utility function that represents the satisfaction level of a portfolio. In determining the utility of a portfolio, the value of wealth and utility function parameters are considered to remain optimal. The results show that the smaller the value of the utility function parameters, the greater the level of satisfaction obtained. Conversely, the greater the wealth value, the lower the investor’s desire to invest. Then, it should be noted that the portfolio model is a stochastic model so that loss valuation needs to be done. Risk measure is one way to perform loss valuation. In this case, the risk measures used are utility function-based expectation, Value-at-Risk (VaR), and Tail Value-at-Risk (TVaR). The three risk measures can be used as a reference in the selection of loss portfolios through stochastic dominance. The results show that the greater the tolerance level of confidence, the greater the VaR and TVaR values obtained. Conversely, for the exponential utility type, the smaller the tolerance level of confidence, the greater the level of satisfaction received. As for the rank utility type, it is found that the smaller the tolerance of the confidence level, the smaller the level of satisfaction received. Therefore, the amount of loss and the level of satisfaction can be taken into consideration in determining the best loss portfolio according to investor preferences.