Moris Simon Strub\'s paper published by Operations Research

2020-04-18

Moris S. Strub Duan Li (2020) Failing to Foresee the Updating of the Reference Point Leads to Time-Inconsistent Investment.

Operations Research 68(1):199-213.

The main question of this paper is whether decision-makers can foresee the update of the reference point so as to make consistent investment in time. The current mainstream literature on behavioral portfolio optimization for reference point update usually assumes that decision makers can foresee the evolution of reference points and thus face the problem of portfolio optimization with consistent time. However empirical studies have shown that decision makers often cannot accurately predict the update of reference points and therefore often make decisions that are not time consistent. In order to further explore this issue this article analyzes and compares five possible risk aversion investment behavior models in discrete time: predetermined static optimal strategy; dynamic optimal strategy under recursive reference point update; dynamic optimal strategy under non-recursive reference point update;  predictable strategy under recursive reference point update; predictable strategy under non-recursive reference point update. By simulating investment behaviors under different frameworks and reference point update rules we found that: in the case of frequent transactions only the dynamic optimal strategy under non-recursive reference point update will positively invest in risky assets over the entire time range. The other four strategies will completely stop investing in risky assets after a short period of time and invest all funds in risk-free assets within the remaining time which is contrary to reality. This finding shows that decision makers cannot foresee the update of the reference point and thus face investments with inconsistent time. Therefore for portfolio research we need to solve the dynamic optimal strategy and update the reference point in a non-recursive manner.

Different from previous studies this study verified the concept of investment behavior that is inconsistent in time due to the unpredictability of reference point updates by simulating investment behavior models under different reference point update rules. The introduction of this concept will not only help the research of portfolio investment theory but also greatly advance the research in other decision science fields.

Link to the paper: https://pubsonline.informs.org/doi/10.1287/opre.2019.1872