Both literature and practice have investigated how the vast amount of ever increasing customer information can inform marketing strategy and decision making. However, the customer data is often susceptible to modeling bias and misleading findings due to various factors including sample selection and unobservable variables. The available analytics toolkit has continued to develop but in the age of nearly perfect information, the customer decision making has also evolved. The dissertation addresses some of the challenges in deriving valid and useful consumer insights from customer data in the digital age. The first study addresses the limitations of traditional customer purchase measures to account of dynamic temporal variations in the customer purchase history. The study proposes a new approach for representation and summarization of customer purchases to improve promotion forecasts. The method also accounts for sample selection bias that arises due to biased selection of customers for the promotion. The second study investigates the impact of increasing internet penetration on the consumer choices and their response to marketing actions. Using the case study of physician’s drug prescribing, the study identifies how marketers can misallocate resources at the regional level by not accounting for variations in internet penetration. The third paper develops a data driven metric for measuring temporal variations in the brand loyalty. Using a network representation of brand and customer the study also investigates the spillover effects of manufacturer related information shocks on the brand’s loyalty.

[pt] Esta dissertação tem como objetivo decompor a desigualdade de rendimentos entre raças e gêneros, em montantes explicáveis pelas escolhas educacional e ocupacional e pela discriminação. Foi utilizada uma metodologia micro- econométrica de simulações contra-factuais, conforme Bourguignon, Ferreira e Lustig (1998), com os dados da PNAD de 1996, incluindo no modelo rendimento mensal, educação, posição na ocupação, entre outras variáveis. Através deste procedimento conseguiu-se atribuir um alto poder explicativo do termo de discriminação, principalmente no caso das mulheres, em que foi computado mais de 100%. Para entender a diferença entre negros e brancos também pode-se recorrer, parcialmente, à desigualdade nas escolhas educacionais, e menos à desigualdade ocupacional. Tendo computado simultaneamente a desigualdade por raça e gênero, nota-se o efeito do gênero sobrepor-se, implicando na perda de poder explicativo da ocupação para o termo de discriminação. Ressalta-se que para homens e mulheres negros a educação dos pais pode ser um importante determinante da desigualdade nos rendimentos. Conclui-se que as políticas sociais voltadas para o combate das desigualdades de rendimentos devem focalizar-se na educação e na transmissão intergeracional, além de procurar medidas que minimizem a diferenciação na remuneração das características. Recomenda-se ainda uma análise temporal, bem como um estudo mais aprofundado sobre os efeitos da educação dos pais. / [en] This thesis is concerned with the analysis of the inequality of earn between races and genders in amounts explained by educational and occupational choices among other characteristics. A micro econometric counterfactual simulation methodology, following Bourguignon, Ferreira e Lustig (1998) was used. The data basis was the PNAD 1996, including monthly earn, education, position, among others. Through this procedure, it was possible to ascertain the high accountability of the discrimination term, mainly referred to women, where over than 100 percent was found. In order to understand the inequality between black and white populations the explanation can be partially referred to the inequality of their educational choices, and also to their occupational choices. For black males and females parents education seems to be a important earn inequality determinant. As a conclusion, social policies concerned with the earn inequalities reduction must aim at education and at the intergenerational transmission of education inequality among blacks, and search for measures that will reduce the differentiation of compensation characteristics. A time series analysis is suggested in order to access a possible decreasing tendency in the term of discrimination, as well as a profound study of the parental education effect on race inequality.