DECISION SCIENCE

Current Projects

GROUP DECISION MAKING WITH MULTIPLE CONFLICTING PERFORMANCE TARGETS

In many multi-stakeholder decision contexts, it is of crucial importance to aggregate preferences in a way that rewards the diversity of targets met. Extending Tsetlin & Winkler’s (2007) work on multi-attribute performance targets, we study the properties of a novel group-weighting rule, which draws on the dis-/similarity of preferences held by individuals and assigns higher values to those alternatives that maximize the diversity of targets achieved.

We then benchmark the performance of this rule against classic approaches based on equal weighting (e.g. average winner, majority voting).

THE EFFECTIVENESS OF TRIMMED OPINION POOLS IN TIME SERIES FORECASTING INVOLVING STRUCTURAL BREAKS

We introduce simple trimming approaches to aggregate judgmental time series forecasts involving structural breaks. While the extant literature explores trimming rules for aggregating forecasts in stable environments, we focus on relatively unstable environments characterized by fundamental regime shifts.

In an empirical study, we find that forecasters are relatively under confident and sensitive to regime shifts and fluctuations in time series, making static trimming approaches less applicable. We further propose asymmetric trimming approaches to aggregate opinion pools under such unstable environments.

HABIT FORMATION IN ENERGY CONSERVATION

A randomized control trial (RCT) targetting 250 residents (with individual air-conditioner units) in the residential colleges of NUS. The objective is to test economic models of habit formation and to evaluate the effectiveness of monetary incentives in inducing a sustainable increase in the household air-conditioner (AC) temperature.

Consistent with theories of habit formation, we find subjects are able to sustain their higher AC temperature post intervention if monetary incentives are paid gradually over a long duration.

THE WISDOM OF CROWDS IN IMPERFECT BINARY MULTI-ATTRIBUTE ENVIRONMENTS

We study the effectiveness of group decisions in binary multi-attribute choice environments when inefficiencies exist among group members regarding the correct evaluation of alternatives.

We focus on the probability that group members using a deterministic elimination-by-aspect (DEBA) heuristic will successfully identify the objectively best alternative in two cases: (1) if there exists uncertainty about the relative importance of attributes and (2) if attribute scores are noisy. We then explore the performance of groups using a majority voting rule to aggregate individual preferences.

Utilizing reduced ordered binary decision diagrams (ROBDDs) to compute exact probabilities of decision performance, we delineate the group’s upper and lower performance boundaries and provide analytical support for the occurrence of crowd wisdom in complex, multidimensional task environments.

ROLE OF SOFT COMMITMENT IN WATER CONSERVATION

A randomized control trial (RCT) targeting 6400 primary school students in Singapore (Grade II to V). The objective is to evaluate and understand the non-monetary soft commitment intervention in the context of water conservation.

The project aims to reduce the school children’s shower time. Data collection completed. Results available.

RISKY CHOICE IN DYNAMIC DECISION ENVIRONMENTS

We study risky choice in dynamic insurance markets. Using a model to integrate prospect theory with system neglect, we predict that in stable environments where the probability of regime shifts is low and individuals observe noisy signals, individuals exhibit risk preferences similar to risk neutral, Bayesian decision-makers in that their difference in willingness-to-pay is not statistically significant.

In unstable environments where the shift probability is high and individuals observe precise signals, risk preferences depend on the type of regime shift. That is, decision-makers are risk seeking when change is dreaded, but exhibit risk aversion when it is desirable.

Importantly, we suggest that DMs risk preferences cannot be attributed to system neglect alone and test our predictions in a laboratory experiment.

RISKY CHOICE FOLLOWING NEAR MISS EVENTS IN SEQUENTIAL TASKS UNDER AMBIGUITY

Studies have shown that near miss events can lead to inconsistent risk perceptions. Yet near misses are often clouded in ambiguity, allowing for hubris and misattribution of what caused success or prevented failure.

We provide an analytical model of near misses and investigate the experience of such an event on risk taking in a real options task.

Over two experiments, we show that increases in risk taking following a near miss occur mainly under ambiguity. We further find that this effect depends on the decision maker’s prior expectation. Only those with an expectation of failure fall prey to the near miss bias.

EXAMINING THE EFFECT OF NEAR MISS EVENTS IN THE HUMANITARIAN NEWSVENDOR CONTEXT

We investigate a humanitarian newsvendor’s pre-positioning decisions in the context of near miss events where the event of interest is a near stock-outs.

Results from a laboratory study show that near stock-outs that are not accompanied by salient warnings regarding the threat of a stock-out lead to normatively inappropriate decreases in pre-positioned stock, corresponding to an increase in risk-taking behavior in future decisions, adding a novel bias to the literature on newsvendor decisions.

Moreover, initial increases in risk-taking following near stock-outs are strongest for decision makers with a predominant focus on avoiding excess inventory as opposed to preventing stock-outs. In the case of repeated experiences of near stock-outs, we witness decreases in risk taking behaviors for near stock-outs with salient cues of the potential danger involved, as well as for those decision makers who primarily focused on excess inventory.

Our study provides clear implications for practice, where unwarranted increases in risk taking behaviors may have disastrous consequences and can easily be prevented by making sure that near miss events regarding supply decisions are properly reported and acknowledged.

VULNERABLE NEAR-MISS EVENTS IN INSURANCE MARKETS: AN EXPERIMENT

Several recent studies have reported changes in risk taking behavior subsequent to experiencing near-miss events. Decision makers observing draws from an (unknown) probability distribution that result in near losses are likely to change their risk perception and thus their valuation of an insurance that could protect them from incurring negative outcomes.

But because insurances are usually traded in imperfect markets and standard Bayesian updating has previously been ruled out as a possible explanation for the near miss bias, we study whether observed changes in supply and demand may result from strategic market uncertainty.

In this study we aim to disentangle these strategic and non-strategic reactions to near-miss events. In our experiment subjects repeatedly face the same gamble with a known probability distribution. We do not focus on probabilistic belief formation but systematically vary the mechanism in which insurances are allocated.

We observe that strategic uncertainty influences demand: Subjects facing human bidders in a call market increase their bids significantly after observing near-loss events while against a random number generator in the BDM mechanism are unaffected. We discuss implications for decision theory and managerial practice.

We use both our own and third-party cookies to enhance our services and to offer you the content that most suits your preferences by analysing your browsing habits. Your continued use of the site means that you accept these cookies. You may change your settings and obtain more information here. Accept