Denoised least squares estimators: An application to estimating advertising effectiveness

It is known in marketing science that an advertiser under- or overspends mi llions of dollars on advertising because the estimation of advertising effe ctiveness is biased. This bias is induced by measurement noise in advertisi ng variables, such as awareness and television rating points, which are pro vided by commercial market research firms based on small-sample surveys of consumers. In this paper, we propose a denoised regression approach to deal with the problem of noisy variables. We show that denoised least squares e stimators are consistent. Simulation results indicate that the denoised reg ression approach outperforms the classical regression approach. A marketing example is presented to illustrate the use of denoised least squares estim ators.