%0 Conference Proceedings
%T Weibull Model for Dynamic Pricing in e-Business
%+ EVF Research Institute, Statistics Department
%+ Applied Mathematics Department
%A Nechval, Nicholas
%A Purgailis, Maris
%A Nechval, Konstantin
%Z Part 7: Innovative e-Business Models
%< avec comité de lecture
%( IFIP Advances in Information and Communication Technology
%B 11th Conference on e-Business, e-Services, and e-Society (I3E)
%C Kaunas, Lithuania
%Y Tomas Skersys
%Y Rimantas Butleris
%Y Lina Nemuraite
%Y Reima Suomi
%I Springer
%3 Building the e-World Ecosystem
%V AICT-353
%P 292-304
%8 2011-10-12
%D 2011
%R 10.1007/978-3-642-27260-8_24
%K e-business
%K pricing
%K uncertainty
%K revenue
%K Weibull model
%K seller risk
%K test plan
%Z Computer Science [cs]
%Z Computer Science [cs]/Networking and Internet Architecture [cs.NI]Conference papers
%X As is the case with traditional markets, the sellers on the Internet do not usually know the demand functions of their customers. However, in such a digital environment, a seller can experiment different prices in order to maximize his profits. In this paper, we develop a dynamic pricing model to solve the pricing problem of a Web-store, where seller sets a fixed price and buyer either accepts or doesn’t buy. Frequent price changes occur due to current market conditions. The model is based on the two-parameter Weibull distribution (indexed by scale and shape parameters), which is used as the underlying distribution of a random variable X representing the amount of revenue received in the specified time period, say, day. In determining (via testing the expected value of X) whether or not the new product selling price c is accepted, one wants the most effective sample size n of observations X1, …, Xn of the random variable X and the test plan for the specified seller risk of Type I (probability of rejecting c which is adequate for the real business situation) and seller risk of Type II (probability of accepting c which is not adequate for the real business situation). Let μ1 be the expected value of X in order to accept c, and μ2 be the expected value of X in order to reject c, where μ1 > μ2, then the test plan has to satisfy the following constraints: (i) Pr{statistically reject c | E{X} = μ1} = α1 (seller risk of Type I), and (ii) Pr{statistically accept c | E{X} = μ2} = α2 (seller risk of Type II). It is assumed that α1 < 0.5 and α2 < 0.5. The cases of product pricing are considered when the shape parameter of the two-parameter Weibull distribution is assumed to be a priori known as well as when it is unknown.
%G English
%Z TC 6
%Z WG 6.11
%2 https://hal.science/hal-01560849/document
%2 https://hal.science/hal-01560849/file/978-3-642-27260-8_24_Chapter.pdf
%L hal-01560849
%U https://hal.science/hal-01560849
%~ IFIP
%~ IFIP-AICT
%~ IFIP-TC
%~ IFIP-WG
%~ IFIP-TC6
%~ IFIP-WG6-11
%~ IFIP-I3E
%~ IFIP-AICT-353