Bayesian Semiparametric Regression: An Exposition and Application to Print Advertising Data

doi:10.1016/S0148-2963(99)00055-7

Journal of Business Research

Volume 49, Issue 3, September 2000, Pages 229-244

https://doi.org/10.1016/S0148-2963(99)00055-7 Get rights and content

Abstract

A new regression-based approach is proposed for modeling marketing databases. The approach is Bayesian and provides a number of significant improvements over current methods. Independent variables can enter into the model in either a parametric or nonparametric manner, significant variables can be identified from a large number of potential regressors, and an appropriate transformation of the dependent variable can be automatically selected from a discrete set of prespecified candidate transformations. All these features are estimated simultaneously and automatically using a Bayesian hierarchical model coupled with a Gibbs sampling scheme. Being Bayesian, it is straightforward to introduce subjective information about the relative importance of each variable, or with regard to a suitable data transformation. The methodology is applied to print advertising Starch data collected from 13 issues of an Australian monthly magazine for women. The empirical results highlight the complex and detailed relationships that can be uncovered using the methodology. An Splus compatible package called “br” that implements the semiparametric approach in this article is currently available at the Statlib repository on the world wide web at http://www.stat.cmu.edu/S/.

Section snippets

Acknowledgements

The authors would like to thank John Rossiter for helping them with the classification in the section titled “Variables Describing the Product Category” and for permission to use the Starch data. Michael Smith was supported by an Australian Postgraduate Research Award and a Monash Faculty grant. The work of Sharat Mathur and Robert Kohn was partially supported by an Australian Research Council Grant. We are grateful to Professors Gary Lilien, David Midgley, John Roberts, and John Rossiter for

References (26)

Michael Smith et al.
Finite Sample Performance of Robust Bayesian Regression
Journal of Computational Statistics
(1996)
H. Akaike
A Bayesian Aanalysis of the Minimum AIC Procedure
Annals Institute Statistical Mathematics
(1978)
George Box et al.
An Analysis of Transformations Revisited, Rebutted
Journal of the American Statistical Association
(1982)
George Box et al.
Bayesian Inference in Statistical Analysis
(1973)
George Casella et al.
Explaining the Gibbs Sampler
The American Statistician
(1992)
David J. Curry
The New Marketing Research SystemsHow to Use Strategic Database Information for Making Better Marketing Decisions
(1993)
Daniel S. Diamond
A Quantitative Approach to Magazine Advertisement Format Selection
Journal of Marketing Research
(1968)
Julia A. Edell et al.
The Information Processing of Pictures in Print Advertising
Journal of Consumer Research
(1983)
Randolf L. Eubank
Spline Smoothing and Nonparametric Regression
(1988)
Adam Finn
Print Ad Recognition Readership ScoresAn Information Processing Perspective
Journal of Marketing Research
(1988)

Lester R. Frankel et al.

Does Recall of an Advertisement Depend upon Its Position in the Magazine?

Journal of Advertising Research

(1963)

Burleig B. Gardner et al.

ROP Color and Its Effect on Newspaper Advertising

Journal of Marketing Research

(1964)

Alan E. Gelfand et al.

Sampling-Based Approaches to Calculating Marginal Densities

Journal of the American Statistical Association

(1990)

Cited by (8)

Print advertising: Executional factors and the RPB Grid
2012, Journal of Business Research
Citation Excerpt :
Indeed, the average full-page magazine ad's Noted score has increased from about 45% several decades ago to 50% now (proprietary Starch INRA Hooper data). The literature establishes the ability of executional factors to increase or depress readership of print ads (Finn, 1988; Rossiter and Percy, 1997; Smith et al., 2000; Franke et al., 2004; Rossiter and Bellman, 2005). The current study selects the executional factors shown by prior research to either increase ad readership or to inhibit ad readership.
This study examines the effects of print ad execution variables on attention and processing as measured by Starch Noted, Associated, and Read most scores—on a univariate basis, then multivariate, and finally within the quadrants of the Rossiter–Percy–Bellman (RPB) Grid. Findings show that executional factors should create high noting scores because initial attention largely determines whether the brand is noticed and its message in the copy read. Univariate and multivariate effects of pictorial and verbal executional factors, ad placement factors, and message style mainly confirm the magazine ad attention tactics in the textbooks by Rossiter and Bellman (2005) and Rossiter and Percy (1997) but with several interesting exceptions. Six of the 16 executional factors investigated differed as a function of the product involvement dichotomy and product purchase motive dichotomy in the RPB Grid. Most (10 of 16), however, apply to all magazine ads, independent of RPB Grid dimensions.
Forecasting television ratings
2011, International Journal of Forecasting
Citation Excerpt :
Hence, we select it to represent the state-of-the-art in forecasting methods. Bayesian model averaging has not previously been used in the context of forecasting television ratings, although it has been employed in the marketing literature to study print advertising (Smith, Mathur, & Kohn, 2000) and to account for the heterogeneity that often arises in the analysis of individual-level marketing data (Gilbride, Allenby, & Brazell, 2006). The covariates used in some of the predictive models fall into two groups, the first being time-based, such as the day of the week, season, year and holidays, and the second being program-specific covariates like genre and duration.
Despite the state of flux in media today, television remains the dominant player globally for advertising spending. Since television advertising time is purchased on the basis of projected future ratings, and ad costs have skyrocketed, there is increasingly pressure to forecast television ratings accurately. The forecasting methods that have been used in the past are not generally very reliable, and many have not been validated; also, even more distressingly, none have been tested in today’s multichannel environment. In this study we compare eight different forecasting models, ranging from a naïve empirical method to a state-of-the-art Bayesian model-averaging method. Our data come from a recent time period, namely 2004–2008, in a market with over 70 channels, making the data more typical of today’s viewing environment. The simple models that are commonly used in industry do not forecast as well as any econometric models. Furthermore, time series methods are not applicable, as many programs are broadcast only once. However, we find that a relatively straightforward random effects regression model often performs as well as more sophisticated Bayesian models in out-of-sample forecasting. Finally, we demonstrate that making improvements in ratings forecasts could save the television industry between $250 and $586 million per year.
Modeling the Level of Drinking Water Clarity in Surabaya City Drinking Water Regional Company Using Combined Estimation of Multivariable Fourier Series and Kernel
2022, Sustainability (Switzerland)
Bayesian variable selection for non-gaussian responses: A marginally calibrated copula approach
2019, arXiv
Smoothing spline in multivariable semiparametric regression through fully Bayesian approach
2018, Model Assisted Statistics and Applications
A systematic review of Bayesian articles in psychology: The last 25 years
2017, Psychological Methods

View all citing articles on Scopus

View full text

Bayesian Semiparametric Regression: An Exposition and Application to Print Advertising Data

Abstract

Section snippets

Acknowledgements

Journal of Computational Statistics

A Bayesian Aanalysis of the Minimum AIC Procedure

Annals Institute Statistical Mathematics

An Analysis of Transformations Revisited, Rebutted

Journal of the American Statistical Association

Bayesian Inference in Statistical Analysis

Explaining the Gibbs Sampler

The American Statistician

The New Marketing Research SystemsHow to Use Strategic Database Information for Making Better Marketing Decisions

A Quantitative Approach to Magazine Advertisement Format Selection

Journal of Marketing Research

The Information Processing of Pictures in Print Advertising

Journal of Consumer Research

Spline Smoothing and Nonparametric Regression

Print Ad Recognition Readership ScoresAn Information Processing Perspective

Journal of Marketing Research

Does Recall of an Advertisement Depend upon Its Position in the Magazine?

Journal of Advertising Research

ROP Color and Its Effect on Newspaper Advertising

Journal of Marketing Research

Sampling-Based Approaches to Calculating Marginal Densities

Journal of the American Statistical Association