The objective of this proposed research is to contribute to the development of a new risky choice theory based on the intuitive idea of mean-risk or risk-value tradeoffs. In this research effort, we will focus on empirical studies of the measure of perceived risk and risk-value theory that we have developed recently.
Intuitively, individuals may consider their choices over risky alternatives based on the idea of a tradeoff between risk and return, where return is typically measured by the mean (or expected return) and risk is measured by some indicator of dispersion or possible losses. This notion is prevalent in the literatures in finance, marketing and other areas. Our risk-value framework provides a new methodology that is totally compatible with this choice behavior. In particular, our approach unifies two streams of research: one in developing preference models and the other in modeling risk judgments. This synthesis makes our risk-value theory more descriptively powerful than other preference models and risk models that have been processed separately for many years.
In this proposed project, we will conduct experimental studies to explore the assumptions and predictions of the measure of perceived risk and of the risk-value models, and evaluate the performance of our models compared with others. These studies will be conducted using different decision problems under different controlled conditions regarding information representation. In order to make our experimental tasks more representative of real choice problems, we will use risky alternatives with a wide range of distributions of multiple outcomes in our studies. Some of our experimental studies will be based on computer programmed decision problems. This will allow subjects to handle experimental tasks and to process information easily, and also let us control experimental conditions and collect data more effectively.
These planned empirical studies should offer significant insights and contribute to our understanding of risk and its role in decision making and risk management. Considering the fact that our models are consistent with a large body of existing empirical evidence regarding both risk and decision making, we expect to receive positive support from our new designed experiments.
