Analytical and Empirical Analyses of Treatment Optimization and Sequential Bargaining: Applications to Opioid Prescriptions and Real Estate Markets

Analytic approaches towards individual decision-making construct a significant portion of the operations management literature. Following its weight in our profession, my doc- toral studies are built upon decision-making problems surrounding two major contexts; opioid prescriptions for pain management in healthcare and price negotiations in real es- tate markets. These studies collaborated with co-authors are presented in this dissertation over three separate chapters. Clinical decisions for opioid prescribing are critical since the prescribing too little can cause patients to suffer from pain whereas prescribing too much may lead to serious drawbacks such as dependence, addition, overdose, and even death. A chapter in this dis- sertation is reserved for a published manuscript collaborated with co-authors, in which, mental trade-offs of clinicians for opioid prescribing decisions are captured in an analyt- ical pain framework. The framework is flexible to capture both acute and chronic pain cases, and yields a minimization objective in terms of opioid prescription duration com- bining the total pain, total discomfort from adverse effects, and the risk of drug inefficacy due to tolerance or increased sensitivity to pain. Despite non-convexity of the objective, a closed-form optimal prescription amount is found. Analyses over the optimal solution show that the role of adverse effects in prescribing decisions is as critical as that of the pain level. Interestingly, we find that the optimal prescription duration is not necessarily increasing with the recovery time. We show that not incorporating the risk of tolerance / increased sensitivity leads to overprescribing. Clinicians’ beliefs on this risk can be cur- tailed at patient handovers, thus leading to overprescribing. We also show overprescrib- ing can be mitigated by adaptive treatments. Lastly, using real-life pain and opioid use data from two sources, we estimate the timing of tolerance / increased pain sensitivity and discuss the proximity of our model to clinical practice. This paper has a pain man- agement framework that leads to tractable models. These models can potentially support balanced opioid prescribing after their validation in a clinical setting. Then, they can be helpful to policymakers in assessment of prescription policies and of the controversy around over(under)prescribing. The other two chapters of the dissertation are reserved for my doctoral studies on price negotiations in real estate markets, both of which are collaborated with co-authors. The mainstream adoption of online marketplaces allows large-scale collection of structured back-and-forth (sequential) bargaining data. Obtaining such data from an online real- estate marketplace company, we empirically analyze sequential price bargains for houses between the company (institutional seller) and individual buyers. In each bargain, par- ties (the seller and buyer) take turns to make concessions until one of them terminates the bargain by accepting the other’s offer or by exiting. Our analyses of concessions and ter- minations respectively yield that parties make diminishing and absolutist (independent of current counteroffer) concessions, whereas they make relativist terminations. These results are robust when tested separately for different buyer and house types, except that certain buyer types make absolutist terminations. We explain relativist (resp. absolutist) acceptances via high (resp. low) sensitivity to the looming deal price, and relativist (resp. absolutist) exits by scarcity (resp. multitude) of available alternatives. Moreover, we ana- lytically show that the empirical properties of concessions apply for compromises as well. These properties of compromises allow us to connect a party’s offers to her reservation price via a simple offer curve. Randomizing offers around the curves, we obtain max- imum likelihood estimates of reservation prices and bargaining powers in terms of the buyer and house types. Providing offer, reservation price, and bargaining power estima- tions, offer curves could lead to a decision support tool for bargaining decisions.