GHGordon Hazen

Gordon Hazen

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Gordon Hazen has a total of 10 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Note: Ordinal Equivalence of Buying and Selling Prices for Information—When is Constant Risk Attitude Entailed?

    by Gordon Hazen

    Hazen and Sounderpandian (1999) claim to have demonstrated that ordinal equivalence of the buying price and the selling price for information entails that the utility function must have constant risk attitude (i.e., be linear or exponential). However, there is no explicit indication of the scope of the ordinal equivalence assumed—is it within any particular decision problem, or within all decision problems sharing the same utility function? Here we explore the difference between these two assumptions, providing counterexamples for the constant risk attitude assertion in the first case, and in the second case a proof that buy-sell ordinal equivalence within all decision problems sharing the same utility forces constant risk attitude. In the first case, the simplest counterexample arises from the assertion that more refined information sources have greater information values.

  • On the Value of Information Across Decision Problems

    by Ali Abbas, Gordon Hazen

    Winner of the 2025 Clemen–Kleinmuntz Decision Analysis Best Paper Award The value of information is an important concept in decision analysis that has been quantified as the buying price (BPI) for the information and the expected utility increase (EUI) obtainable by using the information. These two measures rank information sources identically in a scalar-valued decision problem only when the utility function is linear or exponential. In contrast, this paper focuses on the value of information across scalar-valued decision problems sharing the same utility function such as different divisions within an organization exploring various information sources for their decisions using the same organizational utility function. In this context, it still makes sense to ask which sources are more informative. We show that BPI and EUI rank information sources identically in this context only when the utility function is linear. However, if the certainty equivalent increase is used instead of EUI, then identical ranking with BPI across problems is maintained for the broader class of linear or exponential utility functions. We discuss the importance of these results for distributed decision-making settings, where different departments within an organization may calculate the value of information separately. Our results advise against using EUI to measure information value in this context when risk attitude is important.


  • Information Density in Decision Analysis

    by Gordon Hazen, Emanuele Borgonovo, Xuefei Lu

    Information value has been proposed and used as a probabilistic sensitivity measure, the idea being that uncertain parameters having higher information value are precisely those to which an optimal decision is more sensitive. In this paper, we study the notion of information density as a graphical complement to information value analysis, one that augments an information value calculation with associated directions of information gain. We formally examine mathematical details absent from its earlier presentation that guarantee information density exists and is well posed and describe its relationship to alternate measures of information value. We present its application in the context of a realistic case study and discuss the associated insights.


  • Optimal Enrollment in Late-Stage New Drug Development with Learning of Drug’s Efficacy for Group-Sequential Clinical Trials

    by Zhili Tian, Gordon B. Hazen, Hong Li

    Problem definition: The cost for developing a new drug ranged from $1 billion to more than $2 billion between 2010 and 2019. In addition to high development costs, the efficacy of the candidate drug, patient enrollment, the market exclusivity period (MEP), and the planning horizon are uncertain. Moreover, slow enrollment leads to increased costs, canceled clinical trials, and lost potential revenue. Many firms, hoping to detect efficacy versus futility of the candidate drug early to save development costs, plan interim analyses of patient-response data in their clinical trials. Academic/practical relevance: The problem for optimizing patient-enrollment rates has an uncertain planning horizon. We developed a continuous-time dynamic programming (DP) model with learning of a drug’s efficacy and MEP to assist firms in developing optimal enrollment policies in their clinical trials. We also established the optimality equation for this DP model. Through a clinical trial for testing a cancer drug developed by a leading pharmaceutical firm, we demonstrate that our DP model can help firms effectively manage their trials with a sizable profit gain (as large as $270 million per drug). Firms can also use our model in simulation to select their trial design parameters (e.g., the sample sizes of interim analyses). Methodology: We update a drug’s efficacy by Bayes’ rules. Using the stochastic order and the likelihood-ratio order of distribution functions, we prove the monotonic properties of the value function and an optimal policy. Results: We established that the value of the drug-development project increases as the average response from patients using the candidate drug increases. For drugs having low annual revenue or a strong market brand or treating rare diseases, we also established that the optimal enrollment policy is monotonic in the average patient response. Moreover, the optimal enrollment rate increases as the variance of the MEP decreases. Managerial implications: Firms can use the properties of the value function to select late-stage clinical trials for their drug-development project portfolios. Firms can also use our optimal policy to guide patient recruitment in their clinical trials considering competition from other drugs in the marketplace. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2022.1162 .


  • Augmenting Markov Cohort Analysis to Compute (Co)Variances: Implications for Strength of Cost-Effectiveness

    by Gordon B. Hazen

    Markov cohort analysis is a popular deterministic method in medical decision making for calculating mean outcomes in a Markov model by following a cohort of individuals through time. At present, obtaining outcome variances requires either forsaking cohort analysis in favor of a Markov decision process model or using Monte Carlo simulation (microsimulation), a more computationally demanding procedure that provides only statistical estimates. Here we derive an augmented version of cohort analysis that allows exact computation (not merely estimation) of (co)variances. In second-order models that incorporate parameter uncertainty, augmented cohort analysis can replace the “inner loop” required in Monte Carlo simulation, resulting in quicker and more accurate estimates.One reason for computing variances is to calculate a measure of the strength of an affirmative cost-effectiveness conclusion. In Markov cost-effectiveness analysis, an equivalent measure of cost-effectiveness is positivity of the expected incremental net monetary benefit. Augmented cohort analysis allows calculation of the number of standard deviations that this quantity falls above zero. As a measure of strength of cost-effectiveness, this quantity increases with cohort size. This means that the common practice of taking cohort size to be one can substantially underestimate the strength of a resulting cost-effectiveness conclusion under realistically large cohorts. Moreover, if realistic cohort size is large, then modelers can avoid microsimulation by using augmented cohort analysis and Chebyshev bounds to guarantee the probability of cost effectiveness is close to one.History: Accepted by J. Paul Brooks, Area Editor for Applications in Biology, Medicine, & Healthcare.Supplemental Material: The online appendix is available at https://doi.org/10.1287/ijoc.2022.1234 .


  • Average internal rate of return for risky projects

    by Gordon Hazen, Carlo Alberto Magni

    The average internal rate of return (AIRR) fixes many deficiencies associated with the traditional internal rate of return (IRR), including apparent inconsistency with net present value (NPV). The AIRR approach breaks down project NPV into scale (the capital invested) and economic efficiency (the AIRR), and maintains NPV consistency for accept/reject decisions. Here we examine extensions of the AIRR to risky capital asset projects, a domain where the IRR appears intractable. We show that one can uniquely break down a risky NPV into a risk-sensitive project scale and a risk-sensitive extended AIRR, representing risky project efficiency, so that consistency with NPV for accept/reject decisions is maintained in the certainty-equivalent sense, in direct analogy to the deterministic case. This novel breakdown gives managerial insight by helping determine a risky project’s locus of uncertainty, be it the project scale, or economic efficiency, or both. In this way, risky features of competing projects can be explored in more detail, leading to insights substantiating the NPV ranking. We also show that under risk neutrality, the expected AIRR is equal to the AIRR of the expected cash flow, a property that notoriously fails for the stochastic IRR.


  • Probabilistic sensitivity measures as information value

    by Emanuele Borgonovo, Gordon B. Hazen, Victor Richmond R. Jose, Elmar Plischke


  • ANOVA models for Brownian motion

    by Gordon Hazen, Daniel Apley, Neehar Parikh