東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Envisioning the Improbable: Distribu...

Weston, Shellwyn L.

Linked to FindBook

Google Book

Amazon

博客來

Envisioning the Improbable: Distributional Knowledge and Judgment In Heavy-Tailed Contexts.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Envisioning the Improbable: Distributional Knowledge and Judgment In Heavy-Tailed Contexts./
Author:	Weston, Shellwyn L.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2013,
Description:	87 p.
Notes:	Source: Dissertation Abstracts International, Volume: 75-01(E), Section: A.
Contained By:	Dissertation Abstracts International75-01A(E).
Subject:	Management. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3574123
ISBN:	9781303477737

Envisioning the Improbable: Distributional Knowledge and Judgment In Heavy-Tailed Contexts.
Weston, Shellwyn L.

Envisioning the Improbable: Distributional Knowledge and Judgment In Heavy-Tailed Contexts. - Ann Arbor : ProQuest Dissertations & Theses, 2013 - 87 p.

Source: Dissertation Abstracts International, Volume: 75-01(E), Section: A.

Thesis (Ph.D.)--New York University, Graduate School of Business Administration, 2013.

Heavy-tailed distributions often characterize contexts of great importance to managers (e.g. branded product sales, asset prices, and environmental phenomena) in which lowprobability/high-consequence events occur relatively frequently. Thus, if these contexts are mistakenly characterized as thin-tailed (i.e. contexts where extreme events are exceedingly rare rather than merely unusual), managers may undervalue or dismiss potential blockbuster opportunities, sell assets too cheaply, or fail to plan adequately for catastrophic events. This dissertation highlights the issue that because heavy-tailed phenomena exhibit a much greater than "normal" frequency and size of outliers (probability mass in the tails), they have a much greater than "normal" number of events (probability mass) clustered near the mean. That is, heavy-tailed phenomena often exhibit samples, and sample paths, that appear thin-tailed, owing to the absence of outliers, for large samples or for long periods. This research centers on the question: What judgments do individuals make regarding possible unusual (lowprobability/high-consequence) events in heavy-tailed contexts in the absence of representative experience? The first two experiments demonstrate that individuals overwhelmingly fail to distinguish between heavy- and thin-tailed contexts in the absence of experience. The work then introduces a typology and model of distributional knowledge and the third experiment confirms, using text analysis of individual reasoning statements, that contextual knowledge (the understanding that sample data may be misleading, precipitating a search for analogous contexts, broad categorizations, a generative mechanism, or more data) moderates the biased judgments found in Experiments 1 and 2. The fourth experiment develops and tests a new method, based on financial options theory (Black and Scholes, 1975), of eliciting confidence appropriate for heavy-tailed contexts. Counterintuitively, in the final experiment, individuals with knowledge of at least one distribution in addition to the normal distribution, demonstrate a directionally worse ability to distinguish between heavy- and thin-tailed contexts. This is the first work to develop a typology of distributional knowledge, to model and test individual ability to distinguish between heavy-and thin-tailed contexts, and to employ skewed payoff structures to assess perceptions of tail risk.

ISBN: 9781303477737Subjects--Topical Terms:

516664
Management.

Envisioning the Improbable: Distributional Knowledge and Judgment In Heavy-Tailed Contexts.
LDR:03420nmm a2200301 4500 001 2156316
005 20180517123959.5
008 190424s2013 ||||||||||||||||| ||eng d
020 $a 9781303477737
035 $a (MiAaPQ)AAI3574123
035 $a AAI3574123
040 $a MiAaPQ $c MiAaPQ
100 1 $a Weston, Shellwyn L. $3 3344083
245 1 0 $a Envisioning the Improbable: Distributional Knowledge and Judgment In Heavy-Tailed Contexts.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2013
300 $a 87 p.
500 $a Source: Dissertation Abstracts International, Volume: 75-01(E), Section: A.
500 $a Adviser: Adam M. Brandenburger.
502 $a Thesis (Ph.D.)--New York University, Graduate School of Business Administration, 2013.
520 $a Heavy-tailed distributions often characterize contexts of great importance to managers (e.g. branded product sales, asset prices, and environmental phenomena) in which lowprobability/high-consequence events occur relatively frequently. Thus, if these contexts are mistakenly characterized as thin-tailed (i.e. contexts where extreme events are exceedingly rare rather than merely unusual), managers may undervalue or dismiss potential blockbuster opportunities, sell assets too cheaply, or fail to plan adequately for catastrophic events. This dissertation highlights the issue that because heavy-tailed phenomena exhibit a much greater than "normal" frequency and size of outliers (probability mass in the tails), they have a much greater than "normal" number of events (probability mass) clustered near the mean. That is, heavy-tailed phenomena often exhibit samples, and sample paths, that appear thin-tailed, owing to the absence of outliers, for large samples or for long periods. This research centers on the question: What judgments do individuals make regarding possible unusual (lowprobability/high-consequence) events in heavy-tailed contexts in the absence of representative experience? The first two experiments demonstrate that individuals overwhelmingly fail to distinguish between heavy- and thin-tailed contexts in the absence of experience. The work then introduces a typology and model of distributional knowledge and the third experiment confirms, using text analysis of individual reasoning statements, that contextual knowledge (the understanding that sample data may be misleading, precipitating a search for analogous contexts, broad categorizations, a generative mechanism, or more data) moderates the biased judgments found in Experiments 1 and 2. The fourth experiment develops and tests a new method, based on financial options theory (Black and Scholes, 1975), of eliciting confidence appropriate for heavy-tailed contexts. Counterintuitively, in the final experiment, individuals with knowledge of at least one distribution in addition to the normal distribution, demonstrate a directionally worse ability to distinguish between heavy- and thin-tailed contexts. This is the first work to develop a typology of distributional knowledge, to model and test individual ability to distinguish between heavy-and thin-tailed contexts, and to employ skewed payoff structures to assess perceptions of tail risk.
590 $a School code: 0868.
650 4 $a Management. $3 516664
650 4 $a Behavioral psychology. $3 2122788
650 4 $a Experimental psychology. $3 2144733
690 $a 0454
690 $a 0384
690 $a 0623
710 2 $a New York University, Graduate School of Business Administration. $b Management. $3 3344084
773 0 $t Dissertation Abstracts International $g 75-01A(E).
790 $a 0868
791 $a Ph.D.
792 $a 2013
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3574123