國立虎尾科技大學 |

Operations Strategy on Online Platforms.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Operations Strategy on Online Platforms./
作者:	Chen, Daniel.
面頁冊數:	1 online resource (123 pages)
附註:	Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
Contained By:	Dissertations Abstracts International85-12B.
標題:	Information science. -
電子資源:	click for full text (PQDT)
ISBN:	9798382830049

Operations Strategy on Online Platforms.
Chen, Daniel.

Operations Strategy on Online Platforms. - 1 online resource (123 pages)

Source: Dissertations Abstracts International, Volume: 85-12, Section: B.

Thesis (Ph.D.)--University of Pennsylvania, 2024.

Includes bibliographical references

This dissertation studies operations strategy on online platforms. Platform operations comprise three defining features: the growing use of algorithms and artificial intelligence; large numbers of users with competing interests and incentives for strategic behavior; and large amounts of data. The following three chapters study the design of data-driven policies to improve the operations of online platforms in competitive settings.Chapter 1 studies how gig economy workers make strategic decisions about where and when to work. We empirically measure two types of strategic behavior: multihoming, the choice of an online platform, and repositioning, the choice of a physical location. Using a comprehensive dataset that tracks worker activity across platforms, we develop and estimate a structural model to analyze how workers optimize their earnings by switching between platforms and locations in search of the best trips. We show that workers are highly heterogeneous in their preferences and find multihoming especially costly, both in absolute terms and relative to the cost of repositioning. Through counterfactual simulations, we show that firms and regulators can substantially improve system efficiency by enabling workers to freely multihome, as both workers' hourly earnings and service levels increase in equilibrium. We show that strategic behavior in the gig economy imposes large costs on both individuals and the overall market and that policies to ease the burden of strategic decision-making can significantly improve efficiency.Chapter 2 studies whether operational transparency can improve the impact of algorithmic pricing on customers and workers in the gig economy. We empirically measure the extent of algorithmic bias in pricing on ridehail platforms and model how workers strategically respond to algorithm-generated prices. We show that algorithmic prices vary significantly from any definition of a transparent price that can be explained by observable features of a trip. Despite this variation, we find that algorithmic pricing does not result in bias against low-income locations, as ridehail platforms offer higher service levels and lower prices relative to taxis. However, variation in algorithmic pricing incentivizes workers to decline a large portion of trips in practice, significantly decreasing system efficiency. We estimate policy counterfactuals and show that minimum wage and transparent pricing policies can improve efficiency without harming fairness constraints.Chapter 3 studies methods for using machine learning to generate interpretable causal inference estimates from online datasets. In machine learning settings, prediction errors are a commonly overlooked problem that can bias results and lead to arbitrarily incorrect parameter estimates. We consider a two-stage model where (1) machine learning is used to predict variables of interest, and (2) these predictions are used in a regression model for causal inference. Even when the model specification is otherwise correct, traditional metrics such as p-values and first-stage model accuracy are not good signals of correct second-stage estimates when prediction error exists. We show that these problems are substantial and persist across simulations and an empirical dataset. We propose general sample-based methods to identify when prediction errors are significantly biasing estimates and provide consistent corrections for the case where unbiased training data is available for the machine learning dataset.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798382830049Subjects--Topical Terms:

561178
Information science.
Subjects--Index Terms:

Operations strategyIndex Terms--Genre/Form:

554714
Electronic books.

Operations Strategy on Online Platforms.
LDR:04862ntm a22003977 4500 001 1151716
005 20241118085742.5
006 m o d
007 cr mn ---uuuuu
008 250605s2024 xx obm 000 0 eng d
020 $a 9798382830049
035 $a (MiAaPQ)AAI31143954
035 $a AAI31143954
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Chen, Daniel. $3 1478532
245 1 0 $a Operations Strategy on Online Platforms.
264 0 $c 2024
300 $a 1 online resource (123 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
500 $a Advisor: Allon, Gad.
502 $a Thesis (Ph.D.)--University of Pennsylvania, 2024.
504 $a Includes bibliographical references
520 $a This dissertation studies operations strategy on online platforms. Platform operations comprise three defining features: the growing use of algorithms and artificial intelligence; large numbers of users with competing interests and incentives for strategic behavior; and large amounts of data. The following three chapters study the design of data-driven policies to improve the operations of online platforms in competitive settings.Chapter 1 studies how gig economy workers make strategic decisions about where and when to work. We empirically measure two types of strategic behavior: multihoming, the choice of an online platform, and repositioning, the choice of a physical location. Using a comprehensive dataset that tracks worker activity across platforms, we develop and estimate a structural model to analyze how workers optimize their earnings by switching between platforms and locations in search of the best trips. We show that workers are highly heterogeneous in their preferences and find multihoming especially costly, both in absolute terms and relative to the cost of repositioning. Through counterfactual simulations, we show that firms and regulators can substantially improve system efficiency by enabling workers to freely multihome, as both workers' hourly earnings and service levels increase in equilibrium. We show that strategic behavior in the gig economy imposes large costs on both individuals and the overall market and that policies to ease the burden of strategic decision-making can significantly improve efficiency.Chapter 2 studies whether operational transparency can improve the impact of algorithmic pricing on customers and workers in the gig economy. We empirically measure the extent of algorithmic bias in pricing on ridehail platforms and model how workers strategically respond to algorithm-generated prices. We show that algorithmic prices vary significantly from any definition of a transparent price that can be explained by observable features of a trip. Despite this variation, we find that algorithmic pricing does not result in bias against low-income locations, as ridehail platforms offer higher service levels and lower prices relative to taxis. However, variation in algorithmic pricing incentivizes workers to decline a large portion of trips in practice, significantly decreasing system efficiency. We estimate policy counterfactuals and show that minimum wage and transparent pricing policies can improve efficiency without harming fairness constraints.Chapter 3 studies methods for using machine learning to generate interpretable causal inference estimates from online datasets. In machine learning settings, prediction errors are a commonly overlooked problem that can bias results and lead to arbitrarily incorrect parameter estimates. We consider a two-stage model where (1) machine learning is used to predict variables of interest, and (2) these predictions are used in a regression model for causal inference. Even when the model specification is otherwise correct, traditional metrics such as p-values and first-stage model accuracy are not good signals of correct second-stage estimates when prediction error exists. We show that these problems are substantial and persist across simulations and an empirical dataset. We propose general sample-based methods to identify when prediction errors are significantly biasing estimates and provide consistent corrections for the case where unbiased training data is available for the machine learning dataset.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Information science. $3 561178
653 $a Operations strategy
653 $a Data-driven policies
653 $a Regression model
653 $a Machine learning dataset
655 7 $a Electronic books. $2 local $3 554714
690 $a 0310
690 $a 0796
690 $a 0800
690 $a 0723
710 2 $a University of Pennsylvania. $b Operations, Information and Decisions. $3 1473303
710 2 $a ProQuest Information and Learning Co. $3 1178819
773 0 $t Dissertations Abstracts International $g 85-12B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31143954 $z click for full text (PQDT)