Aristotle University Of Thessaloniki (AUTH)
Στατιστική ανάλυση διάδοσης καινοτομικών προϊόντων σε δικτυακά περιβάλλοντα
Abstract
dc:descriptionIn this doctoral thesis, network construction methods for examining multinational diffusion are proposed and alo the factors which affect the probability of connection within this network structure are examined. Furthermore, the variables which affect time – to – take off are investigated and finally, an agent based model is proposed, in which agents interact within a network environment. The multinational diffusion of innovative products is an interesting field which has been studied by several scholars. In general, the diffusion of innovative products between countries, has been associated with economic and cultural factors, as well as product specific characteristics. The present study aims to investigate if topological structure except for socioeconomic variables affects diffusion of innovative products. The proposed network construction method, is based on the multinational diffusion equation which has been first introduced by Ganesh & Kumar (1996). Network metrics, seem to have a statistically significant effect on the probability the innovative products to diffuse rapidly in the markets, while the effects of the socioeconomic variables are consistent with literature. Furthermore, an alternative way of network construction is proposed based on the paper of Putsis, Balasubramanian, Kaplan, & Sen (1997). A Vector Autoregressive model (VAR) is estimated using the Bass residual series for every pair of countries within each product category. The comparison of the communities created applying the two proposed techniques, revealed a significant degree of overlapping. In addition, the factors which affect the assignment of countries in the produced network communities are investigated. This finding, signals that assignment of countries into communities, is partly ought to the economic and cultural environment of countries, except for network attributes. Finally, an agent based simulation model is proposed, where agents interact with each other within a small – world network topology which evolves though time from complete regular to completely random. Network structure, seems to have a significant effect on the speed of innovative product’s diffusion process within a market.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2017
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Petridis, Nikolaos
- Πετρίδης, Νικόλαος
Subjects
dc:subject × 10Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/41120