c 1. 1 4 [1] 1 CM 850 0003 4 2 1 13.3.7 13.10.10 2 3 4 2. 2.1 2 1 1 1 OR [2] ADBUG 1 42 42 Copyright c by ORSJ. Unauthorized reproduction of this article is prohibited.
[3] [4] SCAN*PRO 2 ROI (Return on Investment) [5] [6] 1 [7] S-O-R Howard Sheth Howard Sheth AIDA (e.g., [8]) (Action) (Attention) (Interest) (Desire) [9] [10] [11] [12] 2.2 1 1 CM 2 SCAN*PRO [5] [13] [14] [15] [16] [17] [18 20] [21] [22] [23] [21] 2 1 1 2014 1 Copyright c by ORSJ. Unauthorized reproduction of this article is prohibited. 43 43
3. CM 3.1 2 1 CM TV-CM KARTE TV-CM KARTE CM GRP CM 1 CM 3 1 1 2 8 1,000 500 500 13 5 4 4 1 2 3 1 3.2 3.2.1 CM CM i y i i =1,...,N CM i M i M i i m X im, W im, Z im X im W im Z im W im Z im W im = 0 Z im= 0 V i L 2 3 (1) M i y i= m=1 p im(x imβ + W imγ + Z imδ)+v i ψ +ε i,ε i N(0,σ 2 ) (1) β, γ, δ ψ p im p im 1 M i pim =1 m=1 p im α m α m α m p im 44 44 Copyright c by ORSJ. Unauthorized reproduction of this article is prohibited.
2 1.000 0.540 1.000 0.333 0.310 1.000 0.472 0.482 0.180 1.000 0.328 0.102 0.157 0.139 1.000 0.312 0.211 0.043 0.236 0.201 1.000 0.332 0.261 0.143 0.120 0.155 0.437 1.000 0.403 0.506 0.291 0.887 0.329 0.304 0.026 1.000 0.302 0.399 0.645 0.140 0.165 0.054 0.168 0.183 1.000 0.069 0.109 0.286 0.410 0.021 1.000 0.168 0.026 0.020 0.336 0.649 0.266 1.000 0.509 0.455 0.241 0.710 0.267 0.392 0.377 1.000 0.157 0.209 0.328 0.066 0.148 0.076 0.096 0.024 1.000 0.4 p im = exp(α m) Mi n=1 exp(αn) (2) α m α 1 >α 2 > >α M M =max{m 1,...,M N} 1 α 1 =0 3.2.2 c CM S c M i y i= m=1 p im(x imβ c + W imγ c + Z imδ c)+v i ψ +ε i,ε i N(0,σ 2 ), if i S c (3) p im β c γ c δ c = λ + ξc, ξc N(0, Ω) (4) 1 3.2.3 MCMC {β, γ,δ} {β c,γ c,δ c} λ α M H (Metropolis Hastings) [24] α p M H 4. 4.1 2007 2 2012 7 2,777 CM 20 CM 36 2014 1 Copyright c by ORSJ. Unauthorized reproduction of this article is prohibited. 45 45
37 CM 1 CM 2010 5 CM A B 2010 2 A B 2 A B B B 1 B 2010 2 2009 8 CM 1 78.8 2 15.7 3 3.5 4 1.4 5 1 10 CM 1 15 30 CM CM CM CM 0 1 (, + ) y i 4 CM 0.242 0.221 0.128 0.769 0.760 0.432 V i GRP 15 4 0 CM CM 0 1 0.006 0.72 CM CM GRP 2,777 5 CM 2,549 228 4.2 3.2.2 2 2 3 4 1 3 3.2.1 3.2.2 3 3 α m =0 2 3.2.1 3.2.2 4 DIC (Deviance Information Criterion) 3 1 2 3 46 46 Copyright c by ORSJ. Unauthorized reproduction of this article is prohibited.
(SSE) {β c,γ c,δ c} λ σ 2 α m = 1 Ω = 0.5m 15,000 10,000 5. 5.1 4 log ML BF BF 0 1 [25] DIC SSE DIC [26] 4 BF [27, 28] SSE 3 1 2 4 log ML log BF DIC SSE 1 1479.8 291.66 2532.1 29.44 2 1446.9 258.73 2458.2 29.29 3 1202.0 13.84 2305.4 25.55 1188.1 2230.1 24.80 5.2 α 5 α HPD [29] α 2 2 2 0.873 2 0.127 2 13 3 β c,γ c,δ c 1 β c,γ c,δ c 2 1 1 1 5 α α HPD 2.5 97.5 1 0 0 0 0 2 1.93 0.48 2.93 1.06 3 2.43 0.47 3.39 1.62 4 2.76 0.48 3.67 1.88 5 3.10 0.50 4.00 2.14 6 3.41 0.48 4.29 2.47 7 3.73 0.46 4.54 2.83 8 4.06 0.42 4.74 3.18 9 4.41 0.34 4.94 3.72 10 4.76 0.25 5.00 4.18 2014 1 Copyright c by ORSJ. Unauthorized reproduction of this article is prohibited. 47 47
1 1 90 HPD 0 90 HPD 0 2 1 3 1.5 90 HPD 90 HPD 6. 6.1 6 2011 8 6 10 90 HPD 5 95 1 0.460 0.349 0.567 2 0.446 0.340 0.565 3 0.433 0.300 0.563 4 0.425 0.336 0.533 5 0.422 0.325 0.519 6 0.421 0.338 0.510 7 0.403 0.308 0.496 8 0.400 0.294 0.522 9 0.398 0.245 0.555 10 0.389 0.304 0.466 0 1 GRP 1500 GRP MCMC (3) MCMC y i HPD 6 48 48 Copyright c by ORSJ. Unauthorized reproduction of this article is prohibited.
6 GRP 1500 GRP 1250 1750 CM 549 0.233 6.2 TV-CM KARTE CM y i CM CM DSP Demand Side Platform CM [30] CM CM 7. CM 2 1 2 2 1 CM CM 1 [1] D. L. Hoffman and T. P. Novak, Marketing Hypermedia Computer-Mediated Environments: Conceptual Foundations, Journal of Marketing, 60, 50 68, 1996. [2] J. D. C. Little, Models and Managers: The Concept of a Decision Calculus, Management Science, 16, B466 B485, 1970. [3] L. G. Cooper and M. Nakanishi, Market-Share Analysis: Evaluating Competitive Marketing Effectiveness, Kluwer Academic Publishers, 1988. [4] D. R. Wittink, M. J. Addona, W. J. Hawkes and J. C. Porter, SCAN*PRO: The Estimation Validation and Use of Promotion Effects Based on Scanner Data, Working Paper of Cornel University, 1988. [5] P. J. Danaher, Advertising Models, in: B. Wieranga (ed.), Handbook of Marketing Decision Models, Ch. 4, 81 106, 2008. [6]M,D.DekimpeandD.M.Hanssens, ThePersistence of Marketing Effects on Sales, Marketing Sci- 2014 1 Copyright c by ORSJ. Unauthorized reproduction of this article is prohibited. 49 49
ence, 14, 1 21, 1995. [7] J. H. Howard and J. N. Sheth, The Theory of Buyer Behavior, John Wily & Sons, 1969. [8] P. Kotler and K. L. Keller, Marketing Management (14th Ed.), Prentice Hall, 2011. [9] SCAP 258, 10 17, 2011 [10] 91 2012 [11] W. W. Moe and P. S. Fader, Dynamic Conversion Behavior at e-commerce Sites, Management Science, 50, 326 335, 2004. [12] R. E. Backlin and C. Sismeiro, A Model of Web Site Browsing Behavior Estimated on Clickstream Data, Journal of Marketing Research, 40, 249 267, 2003. [13] C. Hovland and W. Weiss, The Influence of Source Credibility on Communication Effectiveness, Public Opinion Quarterly, 15, 635 650, 1951. [14] B. Z. Erdogan, Celebrity Endorsement: A Literature Review, Journal of Marketing Management, 15, 291 314, 1999. [15] M. J. Baker and G. A. Churchill, The Impact of Physical Attractive Models on Advertising Evaluations, Journal of Marketing Research, 14, 538 555, 1977. [16] J. Horai, N. Naccari and E. Fatoullah, The Effects of Expertise and Physical Attractiveness upon Opinion Agreement and Liking, Sociometry, 37, 601 606, 1974. [17] R. Ohanian, Construction and Validation of a Scale to Measure Celebrity Endorsers Perceived Expertise, Trustworthiness, and Attractiveness, Journal of Advertising, 19, 39 52, 1990. [18] L. R. Kahle and P. Homer, Physical Attractiveness of the Celebrity Endorser: A Social Adaption Perspective, Journal of Consumer Research, 11, 954 961, 1985. [19] M. A. Kamins and K. Gupta, Congruence between Spokeperson and Product Types: A Matchup Hypothetis Perspective, Psychology and Marketing, 11, 569 586, 1994. [20] M. A. Kamins, An Investigation into the Matchup Hypothesis in Celebrity Advertising: When BeautyMayBeOnlySkinDeep, Journal of Advertising, 19, 4 13, 1990. [21] J. G. Lee and E. Thorson, The Impact of Celebrity-Product Incongruence on the Effectiveness of Product Endorsement, Journal of Advertising Research, 48, 433 449, 2008. [22] G. Mandler, The Structure of Value: Accounting of Taste, In M. S., Clark and S. T. Fisk, (Eds.), Affect and Cognition: The Seventeenth Annual Carnegie Symposium on Cognition, 1982. [23] J. Meyers-Levy, M. and A. M. Tybout, Schema Congruity as a Basis for Product Evaluation, Journal of Consumer Research, 16, 39 54, 1989. [24] M. H. Chen and J. J. Deely, Bayesian Analysis for a Constrained Linear Multiple Regression Problem for Predicting the New Crop of Apples, Journal of Agricultural, Biological and Environmental Statistics, 1, 467 489, 1996. [25] M. A. Newton and A. E. Raftery, Approximate Bayesian Inference with the Weighted Likelihood Bootstrap, Journal of the Royal Statistical Society. Series B, 64, 3 48, 1994. [26] A. Gelman, J. B. Carlin, H. S. Stern and D. B. Rubin, Bayesian Data Analysis, 2nd Edition, Chapman & Hall/CRC, 2004. [27] R. E. Kass and A. E. Raftery, Bayes Factors, Journal of the American Statistical Association, 90, 773 795, 1995. [28] H. Jeffreys, Theory of Probability (3rd Ed.), Oxford: Clarendon Press, 1961. [29] M. H. Chen, Q. M. Shao and J. G. Ibrahim, Monte Carlo Methods in Bayesian Computation, NewYork: Springer, 2000. [30] 2013 2 21 http://www.dentsu.co.jp/news/release/2013/pdf/ 2013016-0221.pdf [31] P. E. Rossi, G. Allenby and R. McCulloch, Bayesian Statistics and Marketing, Wiley, 2005. A. : MCMC [31] (3) Q im =(X im,w im,z im) φ c =(β c,γ c,δ c) φ c K ψ L M i y i = p imq imφ c+v i ψ + εi, ε i N(0,σ 2 ) (5) m=1 MCMC MCMC [22, 23] A.1 π n(x ms 2 ) m s 2 x f(d c θ c) = ( ) Mi π n y i p imq imφ c+v i ψ, σ 2 i Sc m (6) D c c θ c 50 50 Copyright c by ORSJ. Unauthorized reproduction of this article is prohibited.
[ C ] f(θ D) = f(d c θ c)π(φ c λ, Ω) ] (7) π (α m) π(σ 2 )π(λ)π(ω) c=1 [ M m=2 π(α m) = r m(2π) 0.5 exp( 0.5α 2 m)i (α m 1 <α m <α m+1) I(P ) P 1 r m α 1 =0 α M +1= 5 CM ψ N(ψ 0, Ω ψ0) σ 2 Ga(s 0/2,S 0/2) λ N(λ 0, Ω λ0) Ω 1 W (g 0,G 0) Ga( ) W ( ) ψ 0 = 0 Ω ψ0 = 100I L s 0 = S 0 = K λ 0 = 0 Ω λ0 = 100I K g 0 = K G 0 = 100I K A.2 φ c ( φ c N(μ 1S 1) S 1 = σ ( 2 Mi )( ) Mi i Sc m=1 pimqim m=1 pimqim + ) 1 μ1 Ω 1 = S 1 (σ 2 ( M i i Sc m=1 pimq im) ) (y i V i ψ)+ω 1 λ ψ ψ ( N(μ 1S 1) S 1 = σ ) 2 N 1 μ1 i=1 i V i +Ω 1 ψ0 = S ( 1(σ 2 N i=1 i y i ) M i m=1 pimq im +Ω 1 ψ0 ψ 0) α m [ C ] π (α m ) f (D c θ c) π (α m) (8) c=1 M H M H h α N m α h 1 m α N m TN (αm 1,α m+1 )(α h 1 m vα) 2 { } π (α N m a = min 1, π ( α h 1 m ) (9) α h m = α N m α h m = α h 1 m σ 2 σ 2 ( Ga(s 1/2,S 1/2) s 1 = s 0+N S 1 = N ( yi ) ) 1 M 2 i i=1 m=1 pimq imφ c + S 1 0 λ λ N(λ 1, Ω λ1) V λ1 = ( ) CI K +Ω 1 1 λ0 ( λ 1 = V C ) 1 λ1 βc +Ω 1 c=1 λ0 λ 0 Ω Ω 1 W (g 1 (,G 1) g 1 = g 0 + C C ) G 1 = c=1 (φc λ)(φc 1 λ) G 1 0 2014 1 Copyright c by ORSJ. Unauthorized reproduction of this article is prohibited. 51 51