Advertising efficiency based on DEA using the individual-level viewing data

基于DEA的个人观看数据的广告效率

基本信息

批准号：
10430021
负责人：
NAKAJIMA Nozomi
金额：
$ 5.25万
依托单位：
Osaka University
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (B).
财政年份：
1998
资助国家：
日本
起止时间：
1998 至 2000
项目状态：
已结题

项目摘要

This report presents an application of the data envelopment analysis(DEA)to TV advertising planning. The issues of TV advertising planning range from the fundamental measures such as reach and frequency, to more complex problems such as allocating TV commercials to various dayparts.Though much effort has been done on determining advertising-response curves, there is little conclusive evidence as to the shape of these curves. To overcome this difficulty, we formulated a TV advertising planning as an optimization problem with unknown functional form of advertising-response curve and proposed a simple non-parametric approach to the optimization problem with stochastic frontier based on DEA.As a preliminary, we performed a computer simulation for the optimization problems with a Cobb-Douglas and non Cobb-Douglas forms. Simulation results show in each case that an optimal solution to the data envelopment approximation model is sufficiently close to that of the optimization problem with the deterministic part of the stochastic frontier as an objective function, when the size of the observed data is large.Using the awareness data of TV commercials by TV-CM KARTE, allocation problems of commercials to dayparts in various settings are discussed :(1)The allocation of a certain TV commercial to dayparts was solved under the constraint of advertising budget in the case of a priori information on target weights. Sensitivity analysis was also conducted.(2)We employed an interactive approach to the allocation of a certain TV commercial to dayparts in the case of unknown target weights(3)Assuming that the advertiser has a package of spots that are placed in a number of dayparts, instead of budget constraint, we formulated an allocation problem of various TV commercials to dayparts as a multicriteria decision problem. The well-known concept of a compromise solution was used to reconcile conflicting multicriteria.

本报告将数据包络分析（DEA）应用于电视广告计划。电视广告计划的问题范围从诸如覆盖范围和频率之类的基本措施到更复杂的问题，例如将电视广告分配给各种日托。尽管在确定广告响应曲线方面已经付出了很大的努力，但几乎没有任何结论性的证据，关于这些曲线的形状。为了克服这一难度，我们以未知的功能形式的广告响应曲线制定了电视广告计划，作为优化问题，并提出了一种基于DEA的随机边界的简单非参数方法，以实现其优化问题。 Cobb-Douglas和非Cobb-Douglas形式的计算机仿真，以解决优化问题。仿真结果在每种情况下都表明，对于数据包络近似模型的最佳解决方案，当观察到的数据的大小较大时，随机边界的确定性部分作为目标函数，足够接近优化问题的优化问题。讨论了电视CM Karte电视广告的意识数据，讨论了各种环境中商业广告的分配问题：（1）在先验信息的情况下，在广告预算的限制下解决了某些电视广告对Dayparts的分配。在目标重量上。还进行了灵敏度分析。（2）我们采用了一种交互式方法来分配特定的电视广告，以便在未知目标重量的情况下（3）假设广告商有一包斑点，这些景点均放在许多地方Dayparts，而不是预算限制，我们为Dayparts提出了各种电视广告的分配问题，作为多标准决策问题。妥协解决方案的众所周知概念用于调和冲突的多标准。