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

THE DIRICHLET MULTIVARIATE MULTINOMI...

BOYD, MARSHA MICHELE,

Linked to FindBook

Google Book

Amazon

博客來

THE DIRICHLET MULTIVARIATE MULTINOMIAL DISTRIBUTION AS A MAGAZINE EXPOSURE MODEL (ADVERTISING, REACH/FREQUENCY MODELS) /

Record Type:	Electronic resources : Monograph/item
Title/Author:	THE DIRICHLET MULTIVARIATE MULTINOMIAL DISTRIBUTION AS A MAGAZINE EXPOSURE MODEL (ADVERTISING, REACH/FREQUENCY MODELS) // MARSHA MICHELE BOYD.
Author:	BOYD, MARSHA MICHELE,
Description:	1 electronic resource (627 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 46-07, Section: A.
Contained By:	Dissertations Abstracts International46-07A.
Subject:	Mass media. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=8600128
ISBN:	9798661828996

THE DIRICHLET MULTIVARIATE MULTINOMIAL DISTRIBUTION AS A MAGAZINE EXPOSURE MODEL (ADVERTISING, REACH/FREQUENCY MODELS) /
BOYD, MARSHA MICHELE,

THE DIRICHLET MULTIVARIATE MULTINOMIAL DISTRIBUTION AS A MAGAZINE EXPOSURE MODEL (ADVERTISING, REACH/FREQUENCY MODELS) /MARSHA MICHELE BOYD. - 1 electronic resource (627 pages)

Source: Dissertations Abstracts International, Volume: 46-07, Section: A.

This study develops and tests a new multivariate multinomial model for the estimation of advertising vehicle exposure. In addition, eleven other media exposure models are evaluated against a data base of 1,033 tabulated schedules constructed from 1979 Simmons Market Research Bureau data. The types of models tested include: three univariate exposure estimation models; three multivariate models based on the Dirichlet probability distribution; one multivariate multinomial model based on the Dirichlet distribution and the beta matrix expansion method; one simulation model; two approximation models; and two models which employ random estimates of advertising vehicle duplication. The accuracy of the twelve models is assessed via one measure for error in schedule reach estimation and two measures of error in the estimation of the exposure distribution. All models are ranked by relative overall accuracy for the three error measures. Rankings are provided for model performance over the entire data base, and for different schedule sizes and schedule types. For most situations, the approximation models tested provided the most accurate estimates of schedule reach and frequency. These were followed in accuracy by the multivariate models based on the Dirichlet distribution. The univariate models were next most accurate, followed by the simulation model, then the random duplication models. The proposed new multivariate multinomial model was generally no more accurate than the univariate models for the estimation of reach. For estimation of the exposure distribution, the model proved more accurate than the univariate models but less accurate than the Dirichlet multivariate models. The findings on relative model performance are congruent with existing theory and previous research, except for the disappointing performance of the multivariate multinomial model. And, the application of the new model was found to be impractical for magazine schedules containing more than seven vehicles. The results contradict the assumption that increased accuracy is gained through preservation of individual vehicle data. Further investigation is warranted to explain this discrepancy. Two issues related to reach/frequency estimation are examined in depth: (1) various approaches to the estimation of vehicle duplication; and (2) the need for uniform reach/frequency model evaluation standards.

English

ISBN: 9798661828996Subjects--Topical Terms:

516793
Mass media.

THE DIRICHLET MULTIVARIATE MULTINOMIAL DISTRIBUTION AS A MAGAZINE EXPOSURE MODEL (ADVERTISING, REACH/FREQUENCY MODELS) /
LDR:03601nmm a22003373i 4500 001 2434414
005 20251027113109.5
006 m o d
007 cr|nu||||||||
008 261011s1985 miu||||||m |||||||eng d
020 $a 9798661828996
035 $a (MiAaPQD)AAI8600128
035 $a AAI8600128
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a BOYD, MARSHA MICHELE, $e author. $3 3824395
245 1 4 $a THE DIRICHLET MULTIVARIATE MULTINOMIAL DISTRIBUTION AS A MAGAZINE EXPOSURE MODEL (ADVERTISING, REACH/FREQUENCY MODELS) / $c MARSHA MICHELE BOYD.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 1985
300 $a 1 electronic resource (627 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: 46-07, Section: A.
500 $a Publisher info.: Dissertation/Thesis.
502 $b Ph.D. $c University of Illinois at Urbana-Champaign $d 1985.
520 $a This study develops and tests a new multivariate multinomial model for the estimation of advertising vehicle exposure. In addition, eleven other media exposure models are evaluated against a data base of 1,033 tabulated schedules constructed from 1979 Simmons Market Research Bureau data. The types of models tested include: three univariate exposure estimation models; three multivariate models based on the Dirichlet probability distribution; one multivariate multinomial model based on the Dirichlet distribution and the beta matrix expansion method; one simulation model; two approximation models; and two models which employ random estimates of advertising vehicle duplication. The accuracy of the twelve models is assessed via one measure for error in schedule reach estimation and two measures of error in the estimation of the exposure distribution. All models are ranked by relative overall accuracy for the three error measures. Rankings are provided for model performance over the entire data base, and for different schedule sizes and schedule types. For most situations, the approximation models tested provided the most accurate estimates of schedule reach and frequency. These were followed in accuracy by the multivariate models based on the Dirichlet distribution. The univariate models were next most accurate, followed by the simulation model, then the random duplication models. The proposed new multivariate multinomial model was generally no more accurate than the univariate models for the estimation of reach. For estimation of the exposure distribution, the model proved more accurate than the univariate models but less accurate than the Dirichlet multivariate models. The findings on relative model performance are congruent with existing theory and previous research, except for the disappointing performance of the multivariate multinomial model. And, the application of the new model was found to be impractical for magazine schedules containing more than seven vehicles. The results contradict the assumption that increased accuracy is gained through preservation of individual vehicle data. Further investigation is warranted to explain this discrepancy. Two issues related to reach/frequency estimation are examined in depth: (1) various approaches to the estimation of vehicle duplication; and (2) the need for uniform reach/frequency model evaluation standards.
546 $a English
590 $a School code: 0090
650 4 $a Mass media. $3 516793
650 4 $a Mass communications. $3 3422380
690 $a 0708
710 2 $a University of Illinois at Urbana-Champaign. $e degree granting institution. $3 3765901
773 0 $t Dissertations Abstracts International $g 46-07A.
790 $a 0090
791 $a Ph.D.
792 $a 1985
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=8600128