Chasing Demand: Learning and Earning in a Changing Environment

Keskin, N. Bora; Zeevi, Assaf

doi:10.2139/ssrn.2389750

Cited by 36 publications

(65 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Comparison with antecedent work on learning in changing environments. In a recent study, Keskin and Zeevi (2016) analyzed the performance of various forced-exploration dynamic pricing policies in changing environments. In particular, they constructed and studied policies that employ moving windows, gradually decaying weights, and change-point detection tests.…”

Section: Discussionmentioning

confidence: 99%

“…Our work differs from theirs in two major ways. First, the focus of Keskin and Zeevi (2016) is on forced exploration; all of their policies rely on a pre-determined schedule of explicit experiments with controls, rather than using certainty-equivalence. By contrast, our work studies the performance of certainty-equivalence policies without any forced exploration.…”

Section: Discussionmentioning

confidence: 99%

“…By contrast, our work studies the performance of certainty-equivalence policies without any forced exploration. Secondly, the aforementioned moving windows in Keskin and Zeevi (2016) are chosen deterministically in the beginning of the time horizon, whereas the estimation windows of the policy C * are chosen adaptively. Consequently, the absence of forced exploration and the adaptive nature of estimation windows make our policies and their analysis distinct.…”

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

On Incomplete Learning and Certainty-Equivalence Control

Keskin

Zeevi

2018

Operations Research

Self Cite

View full text Add to dashboard Cite

We consider a dynamic learning problem where a decision maker sequentially selects a control and observes a response variable that depends on chosen control and an unknown sensitivity parameter. After every observation, the decision maker updates her/his estimate of the unknown parameter and uses a certainty-equivalence decision rule to determine subsequent controls based on this estimate. We show that under this certainty-equivalence learning policy the parameter estimates converge with positive probability to an uninformative fixed point that can differ from the true value of the unknown parameter; a phenomenon that will be referred to as incomplete learning. In stark contrast, it will be shown that this certainty-equivalence policy may avoid incomplete learning if the parameter value of interest "drifts away" from the uninformative fixed point at a critical rate. Finally, we prove that one can adaptively limit the learning memory to improve the accuracy of the certainty-equivalence policy in both static (estimation), as well as slowly varying (tracking) environments, without relying on forced exploration.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

On Incomplete Learning and Certainty-Equivalence Control

Keskin

Zeevi

2018

Operations Research

Self Cite

View full text Add to dashboard Cite

show abstract

“…因此, 一个研究问题是希望设计的算法可以做到较强的鲁棒性 (robustness), 即当外界环境可能遇到一定干扰 (但是有限的干扰) 时, 此算法还能保证有较好的表现. 这方面已有少量研究 (如文献 [103,104]), 但有更多的问题等待人们去进行探索.…”

Section: 关于库存问题的进一步论述可以参考后文相关章节unclassified

Modern optimization theory and applications

Deng¹,

Jian-jun²,

Ge³

et al. 2020

Sci. Sin.-Math.

View full text Add to dashboard Cite

线性规划问题可以说是目前优化领域中最重要且已经被大量研究过的一类问题. 本节将回顾 3 个线性规划相关主题的最新进展: (i) 线性规划算法; (ii) 线性规划和 Markov 决策过程; (iii) 在线线性规划. 本节主要介绍 (i) 和 (ii) 两个领域近期一些比较重要和有趣的发现. (iii) 将在下一节讨论, 希望这些讨论能激发科研工作者在这些领域进行新的研究.

show abstract

“…Before running the experiment we simulate outcomes we could obtained from a field test based on assumed parameters. The tradeoff between exploration and exploitation was a big part of our conversations when working on part of the project both internally within the research team and when interacting with our retail partner (for a detail discussion on the exploration and exploitation issues see Sauré and Zeevi (2013), Keskin and Zeevi (2013) and Keskin and Zeevi (2014)). We estimated, to the best of our knowledge, what were minimum bounds in terms of variation and duration of the test that we needed for the field test to be effective.…”

Section: Field Experiments Design and Implementationmentioning

confidence: 99%

Competition-Based Dynamic Pricing in Online Retailing: A Methodology Validated with Field Experiments

Fisher

Gallino

2015

SSRN Journal

View full text Add to dashboard Cite

A retailer following a competition-based dynamic pricing strategy tracks competitors' price changes and then must answer the following questions: (1) Should the retailer respond? (2) If so, respond to whom? (3) How much of a response? (4) And on which products? The answers require unbiased measures of price elasticity as well as accurate knowledge of competitor significance and the extent to which consumers compare prices across retailers. To quantify these factors empirically, there are two key challenges: first, the endogeneity associated with almost any type of observational data, where prices are correlated with demand shocks observable to pricing managers but not to researchers; and second, the absence of competitor sales information, which prevents efficient estimation of a full consumer-choice model. We address the first issue by conducting a field experiment with randomized prices. We resolve the second issue by proposing an identification strategy that exploits the retailer's own and competitors' stock-outs as a valid source of variation to the consumer choice set. We estimate an empirical model capturing consumer choices among substitutable products from multiple retailers. Based on the estimates, we propose a best-response pricing strategy that takes into account consumer choice behavior, competitors' actions, and supply parameters (procurement costs, margin target, and manufacturer price restrictions). We test our algorithm through a carefully controlled live experiment that lasts for five weeks. The experiment documents an 11 percent revenue increase, while maintaining margin above a retailer specified target.

show abstract

Chasing Demand: Learning and Earning in a Changing Environment

Cited by 36 publications

References 29 publications

On Incomplete Learning and Certainty-Equivalence Control

On Incomplete Learning and Certainty-Equivalence Control

Modern optimization theory and applications

Competition-Based Dynamic Pricing in Online Retailing: A Methodology Validated with Field Experiments

Contact Info

Product

Resources

About