東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Essays on social networks and new pr...

Van den Bulte, Christophe.

FindBook

Google Book

Amazon

博客來

Essays on social networks and new product diffusion.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Essays on social networks and new product diffusion./
作者:	Van den Bulte, Christophe.
面頁冊數:	208 p.
附註:	Source: Dissertation Abstracts International, Volume: 59-06, Section: A, page: 2108.
Contained By:	Dissertation Abstracts International59-06A.
標題:	Business Administration, Marketing. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9836784
ISBN:	0591903970

Essays on social networks and new product diffusion.
Van den Bulte, Christophe.

Essays on social networks and new product diffusion. - 208 p.

Source: Dissertation Abstracts International, Volume: 59-06, Section: A, page: 2108.

Thesis (Ph.D.)--The Pennsylvania State University, 1998.

This dissertation consists of three essays. The first essay assesses the effect of R&D team co-location on communication patterns among R&D and marketing. It applies log-linear network modeling and QAP matrix regression to network data collected from a quasi-experiment at a company that regrouped its R&D teams into a new facility but away from marketing. The co-location increased the communication frequency among R&D teams but did not decrease the communication frequency between R&D and marketing, suggesting that the effect of co-location depends on the content of the communication flows.

ISBN: 0591903970Subjects--Topical Terms:

1017573
Business Administration, Marketing.

Essays on social networks and new product diffusion.
LDR:03181nmm 2200313 4500 001 1840170
005 20050714101549.5
008 130614s1998 eng d
020 $a 0591903970
035 $a (UnM)AAI9836784
035 $a AAI9836784
040 $a UnM $c UnM
100 1 $a Van den Bulte, Christophe. $3 1928517
245 1 0 $a Essays on social networks and new product diffusion.
300 $a 208 p.
500 $a Source: Dissertation Abstracts International, Volume: 59-06, Section: A, page: 2108.
500 $a Adviser: Gary L. Lilien.
502 $a Thesis (Ph.D.)--The Pennsylvania State University, 1998.
520 $a This dissertation consists of three essays. The first essay assesses the effect of R&D team co-location on communication patterns among R&D and marketing. It applies log-linear network modeling and QAP matrix regression to network data collected from a quasi-experiment at a company that regrouped its R&D teams into a new facility but away from marketing. The co-location increased the communication frequency among R&D teams but did not decrease the communication frequency between R&D and marketing, suggesting that the effect of co-location depends on the content of the communication flows.
520 $a The second essay presents a family of micro-level hazard models in which the effect of social networks on new product adoption can be captured explicitly and assessed empirically. Two key ideas are to (1) construct network autocorrelation variables to represent the amount of incentive or pressure that an individual experiences to adopt, and (2) use discrete-time hazard functions that can be vested in random utility theory. Applying the framework to the Medical Innovation data (Coleman et al. 1966) indicates that, contrary to the traditional interpretation, the diffusion of tetracycline does not contain compelling evidence for the existence of social network effects. Rather, marketing effort is at least as likely to have driven the diffusion process.
520 $a The third essay shows that the nonlinear least squares estimates of the Bass diffusion model are biased, and change systematically as one extends the number of observations used in the estimation, even in the absence of model misspecification. The amount of bias is a function of the amount of noise in the data, the number of data points, and the difference between the cumulative penetration observed and the true penetration ceiling (i.e., the extent of right censoring). In situations consistent with marketing applications, m and p can be underestimated by 20% and q overestimated by 30%. The existence of biases that become smaller as the number of data points increases and the censoring decreases explains why systematic changes in the parameter estimates are often observed. Also, extant evidence that diffusion patterns have accelerated over time can be explained as merely a method artifact.
590 $a School code: 0176.
650 4 $a Business Administration, Marketing. $3 1017573
650 4 $a Sociology, Theory and Methods. $3 626625
650 4 $a Engineering, Industrial. $3 626639
690 $a 0338
690 $a 0344
690 $a 0546
710 2 0 $a The Pennsylvania State University. $3 699896
773 0 $t Dissertation Abstracts International $g 59-06A.
790 1 0 $a Lilien, Gary L., $e advisor
790 $a 0176
791 $a Ph.D.
792 $a 1998
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9836784