No description available.

Dada a crescente importância do relacionamento empresa-cliente e a pluralidade de modelos que discutam os elementos e resultados desse tipo de atividade, esta pesquisa buscou entender melhor a atividade de marketing de relacionamento e seus resultados por meio da análise de como os profissionais dessa área enxergam as relações entre os constructos relevantes, a fim de buscar desenvolver um modelo que expresse mais claramente como os profissionais da área entendem que os resultados de marketing de relacionamento são alcançados. O tema é tratado com base em uma revisão teórica que explora o conceito de marketing de relacionamento e seus principais elementos, o processo de modelagem, e o processo de mapeamento cognitivo. A partir da revisão teórica, foram listados onze constructos de marketing de relacionamento, que foram, então, analisados por 35 profissionais da área. Assim foram traçados 35 mapas cognitivos individuais que apontam como cada entrevistado entende as interações entre os elementos do marketing de relacionamento. Com base nas visões de cada indivíduo, foi traçado um mapa cognitivo agregado que aponta o entendimento médio dos entrevistados. A análise desse mapa permitiu identificar como, de acordo com a visão dos profissionais pesquisados, os elementos de marketing de relacionamento se influenciam mutuamente e quais elementos têm maior ou menor impacto no resultado final (aqui considerado como o custumer lifetime value - CLV). Por fim, foi feita análise de cluster para identificar a possível existência de algum tipo de agrupamento de respostas. Nenhum padrão significativo foi encontrado / Given the growing importance of activities related to the relationship between company and client, and the plurality of models to discuss the elements and results of this type of activity, better understanding the relationship marketing activity and it\'s results is important, in order to seek to develop a model that express more clearly how relationship marketing results are achieved. The theme is treated from a theoretical review based on the concept of relationship marketing and its key elements, the modeling process, and the process of cognitive mapping. Eleven contructos of relationship were listed from the theoretical review, and were then analyzed by 35 marketing professionals. Thus 35 individual maps which indicate how each respondent understands the interactions between the elements of relationship marketing have been traced. Based on the views of each individual, was plotted an aggregate map that shows the average understanding of the respondents. The analysis of this map has allowed us to identify, in accordance with the vision of

Cette thèse traite de la compréhension de l’engagement client dans un contexte multicanal. Le flot de littérature sur le marketing multicanal présente plusieurs lacunes théoriques et managériales, telles que les impacts dynamiques de l’adoption multicanal sur la valeur client, la quantification de la rentabilité des campagnes marketing, le comportement multicanal à travers les marques, etc. (Neslin et al. 2014; Neslin and Shankar 2009). Cette thèse a pour but d’éclairer les impacts dynamiques de l’engagement client sur la valeur client et les revenus de l’entreprise dans un contexte multicanal. Le Chapitre 1 aborde les « Impacts dynamiques du canal d’acquisition et de l’adoption multicanal sur la valeur vie client ». Au Chapitre 2, « antécédents et conséquences de l’utilisation des codes promotionnels », nous étudions le processus d’utilisation des codes promotionnels et la rentabilité de telles campagnes. Au Chapitre 3, « Modéliser les impacts de l’achat multicanal sur le choix de marque », nous étudions la question du comportement en termes de choix d’une marque dans un environnement multicanal. Dans l’ensemble, cette thèse étudie divers engagements client et leurs effets sur la valeur client et les revenus de l’entreprise. Du point de vue théorique, c’est une contribution aux publications sur le marketing dans les domaines du marketing multicanal, de la valeur vie client, des promotions et des choix de marques. Elle propose une approche exhaustive de l’engagement client et de la valeur client dans un contexte multicanal. Dans une optique de gestion, cette étude propose aux entreprises des méthodologies novatrices pour gérer leur clientèle au niveau individuel, ainsi que de nouveaux modèles pour évaluer les activités de marketing multicanal / This dissertation aims to investigate the dynamic impacts of customer engagement on customer value and firm revenues in a multichannel context. I address this research question in three chapters. In chapter 1, we propose a hidden Markov model to understand the dynamic effects of acquisition channel and multichannel adoption on the customer-firm relationship and to estimate the customer lifetime value. The results show that multichannel customers acquired from offline channels exhibit higher short-term value than multichannel customers acquired from online channels. In contrast, multichannel customers acquired from online channels are more likely to stay in a higher value state in the long run. In the long run, multichannel customers acquired from offline channels are more valuable than other customers. Chapter 2 focuses on promo code redemption behaviors. We conduct the research using a field experiment. We identify the determinants of opening, clicking the email and the final redemption behavior using a simultaneous multiple equation probit model. In our setting, we find that 11% of the profitability of the promo code campaign stems from eligible purchases without redemption. By targeting those customer segments that are more likely to make eligible purchases without redemption, the average profitability increases. In Chapter 3, we investigate how multichannel behavior impacts brand choice in a grocery setting. Our research develops a hierarchal Bayesian brand choice decision model to understand how multichannel adoption impacts brand choice decision, brand size-of-wallet, and brand share-of-wallet. Our results show that brand choice probability increases in general after consumers become multichannel. We also find that consumers increase both the size-of-wallet and share-of-wallet for the brand which they first purchase online, yet do not necessarily decrease the share-of-wallet for other brands.

Prior researches in P2P file sharing mostly focus on several topics such as the overlay topology, the content searching, the peer discovery, the sharing fairness, incentive mechanisms, except scheduling algorithms for peer-to-peer collaborative file distribution. The scheduling algorithm specifies how file pieces are distributed among peers. When a peer that has the rarest piece leaves, the other peers probably download the incomplete file in the network. Our algorithm is involved in the lifetime of peers in the P2P networks. We first use the distribution of peer¡¦s lifetime and the demand of each peer to decide which peers send which pieces that are rarities, and then also consider the distribution of peer¡¦s lifetime and the demand of each peer to decide which peers receive these pieces. Our goals are to maximize number of peers which have downloaded an entire file before it leaves, to increase the availability of different file pieces, and to minimize the transmission time of the latest completion. Lastly, we show the comparison of the performances of RPF, MDNF, Lifetime-based RPF, and Lifetime-based MDNF algorithms.

published_or_final_version / Architecture / Master / Master of Architecture

The reported lifetime in an in-beam neutron lifetime experiment performed at NIST was tn = (886.3 ± 3.4) s. The largest source of uncertainty was the efficiency of the neutron flux monitor (0.3% relative uncertainty). The flux monitor operates by counting charged particles produced when neutrons impinge on a 6Li foil. Its efficiency was calculated from the 6Li thermal neutron cross section, the solid angle subtended by the charged particle detectors, and the amount of neutron-absorbing material present on the foil. An absolute black neutron detector for cold neutron beams has been developed to measure the efficiency without the need to know these quantities. The flux monitor efficiency is measured to a precision of 0.052% using this direct calibration technique. This calibration removes the largest barrier to a 1 s neutron lifetime measurement with the beam technique. It is hoped that this data can also be used to re-evaluate the current NIST neutron lifetime value, reduce its uncertainty, and remove the dependence on evaluated nuclear data files. There is also the possibility for a direct measurement of the 6Li thermal neutron cross section.

Dada a crescente importância do relacionamento empresa-cliente e a pluralidade de modelos que discutam os elementos e resultados desse tipo de atividade, esta pesquisa buscou entender melhor a atividade de marketing de relacionamento e seus resultados por meio da análise de como os profissionais dessa área enxergam as relações entre os constructos relevantes, a fim de buscar desenvolver um modelo que expresse mais claramente como os profissionais da área entendem que os resultados de marketing de relacionamento são alcançados. O tema é tratado com base em uma revisão teórica que explora o conceito de marketing de relacionamento e seus principais elementos, o processo de modelagem, e o processo de mapeamento cognitivo. A partir da revisão teórica, foram listados onze constructos de marketing de relacionamento, que foram, então, analisados por 35 profissionais da área. Assim foram traçados 35 mapas cognitivos individuais que apontam como cada entrevistado entende as interações entre os elementos do marketing de relacionamento. Com base nas visões de cada indivíduo, foi traçado um mapa cognitivo agregado que aponta o entendimento médio dos entrevistados. A análise desse mapa permitiu identificar como, de acordo com a visão dos profissionais pesquisados, os elementos de marketing de relacionamento se influenciam mutuamente e quais elementos têm maior ou menor impacto no resultado final (aqui considerado como o custumer lifetime value - CLV). Por fim, foi feita análise de cluster para identificar a possível existência de algum tipo de agrupamento de respostas. Nenhum padrão significativo foi encontrado / Given the growing importance of activities related to the relationship between company and client, and the plurality of models to discuss the elements and results of this type of activity, better understanding the relationship marketing activity and it\'s results is important, in order to seek to develop a model that express more clearly how relationship marketing results are achieved. The theme is treated from a theoretical review based on the concept of relationship marketing and its key elements, the modeling process, and the process of cognitive mapping. Eleven contructos of relationship were listed from the theoretical review, and were then analyzed by 35 marketing professionals. Thus 35 individual maps which indicate how each respondent understands the interactions between the elements of relationship marketing have been traced. Based on the views of each individual, was plotted an aggregate map that shows the average understanding of the respondents. The analysis of this map has allowed us to identify, in accordance with the vision of

Submitted by Débora Cabral Soares Ribeiro (d.c.soares@uol.com.br) on 2017-01-16T00:56:39Z No. of bitstreams: 1 Dissertação_Biblioteca_Debora_Ribeiro.pdf: 3530018 bytes, checksum: aa769311a68fa5ddac2f33dc9697d7ea (MD5) / Rejected by Fabiana da Silva Segura (fabiana.segura@fgv.br), reason: Boa Tarde, Prezada Débora O titulo do trabalho está diferente do título que foi aprovado em ATA, seu orientador deverá encaminhar e-mail para o mestradoprofissional@fgv.br mencionando o novo título, somente após o envio do e-mail você poderá postar novamente A ordem das páginas são: Capa - Contra Capa, ficha catalográfica e folha de assinaturas Após o envio do e-mail do seu orientador e a correção da ordem dessas primeiras páginas submeter novamente o trabalho on 2017-01-16T15:18:29Z (GMT) / Submitted by Débora Cabral Soares Ribeiro (d.c.soares@uol.com.br) on 2017-01-18T18:04:57Z No. of bitstreams: 1 Dissertação_Biblioteca_Debora_Ribeiro.pdf: 3529746 bytes, checksum: 6177b546ae6848a69923478fbe385136 (MD5) / Approved for entry into archive by Fabiana da Silva Segura (fabiana.segura@fgv.br) on 2017-01-18T18:11:24Z (GMT) No. of bitstreams: 1 Dissertação_Biblioteca_Debora_Ribeiro.pdf: 3529746 bytes, checksum: 6177b546ae6848a69923478fbe385136 (MD5) / Made available in DSpace on 2017-01-19T11:42:04Z (GMT). No. of bitstreams: 1 Dissertação_Biblioteca_Debora_Ribeiro.pdf: 3529746 bytes, checksum: 6177b546ae6848a69923478fbe385136 (MD5) Previous issue date: 2016-12-20 / The customer lifetime value (CLV) allows to measure the present value of the future cash flows in the relationship of a company with its client. The CLV calculation includes retention and acquisitions analysis, retention and acquisitions expenses, margin obtained by the company with the customer, media expenses, and other investments made by the company at the customer. Several applications have been proposed for CLV, ranging from companies in business-tobusiness and business markets to consumer companies. In addition, there are studies that use the CLV assumptions to measure the value of the sales agent. In the context of direct selling, the direct seller / consultant has an essential role for the business, since in addition to selling, he is also a consumer, using for his own consumption. Usually the direct seller works for more than one brand. Withholding is a challenge because in addition to losing a customer, you may be losing a seller to competing companies. Direct sales companies make high retention expenses with high promotional incentives. Thus, there are opportunities for the application of CLV in direct sales, contributing to the targeting of these incentives in order to maintain the profitability of the business. This project addresses the CLV metrics, specifically applied to analyze the value of this profile of agent that at the same time is a customer and sales representative: the consultant of direct selling products, more specifically in the personal hygiene, perfumery and (HPPC) in Brazil, given the importance of direct sales to HPPC. This question is relevant since, when the customer value (CLV) is identified, it is possible, for example, to determine the investment allowed, increasing or limiting, according to the expected return. The objective of this work is to find the value of the consultant. In order to achieve this objective, the data of 19827 consultants were analyzed, evaluating cadastral, transactional information, shopping baskets and response to stimuli, considered a period of 28 historical cycles, equivalent to just over 1.5 years. The analysis consisted in evaluating the profitability of this agent in a future period equivalent to 6 cycles, about 4 months, from the probability of this consultant being active in each one of the periods. It is possible to identify groups of consultants who may receive different investments, according to the profitability range of each group. / O valor do cliente no tempo, ou customer lifetime value (CLV) permite mensurar o valor presente dos fluxos de caixa futuro no relacionamento de uma empresa com seu cliente. O cálculo do CLV compreende análise de retenção, gastos em retenção, análise da aquisição, gastos em aquisição, margem obtida pela empresa com o cliente, gastos em mídias, e outros investimentos feitos pela empresa no cliente. Diversas aplicações têm sido propostas para o CLV, desde empresas em mercados empresariais business-to-business até empresas de consumo. Inclusive, há estudos que utilizam as premissas do CLV para mensurar o valor do agente de vendas. No contexto da venda direta, o revendedor possui um papel essencial para o negócio, uma vez que além de vender, ele também é consumidor, utilizando para consumo próprio. Normalmente o revendedor atua para mais de uma marca. A retenção é um desafio pois além de perder um cliente, pode estar perdendo um vendedor para empresas concorrentes. As empresas de vendas diretas fazem elevados gastos de retenção, com elevados incentivos promocionais. Assim, há oportunidades para aplicação do CLV na venda direta, contribuindo para o direcionamento desses incentivos de forma a manter a lucratividade do negócio. Este projeto aborda a métrica do CLV, especificamente, aplicado para analisar o valor deste perfil de agente que ao mesmo tempo é cliente e representante de vendas: o consultor/revendedor de produtos de venda direta, mais especificamente no mercado de higiene pessoal, perfumaria e cosméticos (HPPC) no Brasil, dada a importância da venda direta para o HPPC. Esta questão é relevante uma vez que ao ser identificado o valor do cliente (CLV) é possível, por exemplo, determinar o investimento permitido, aumentando ou limitando, de acordo com o retorno esperado. O objetivo deste trabalho é encontrar o valor do consultor. Para atingir este objetivo foram analisados os dados de 19827 consultores, avaliando informações cadastrais, transacionais, cestas de compra e resposta à estímulos, considerado um período de 28 ciclos históricos, o equivalente a pouco mais de 1 ano e meio. A análise consistiu em avaliar a lucratividade deste agente em um período futuro equivalente a 6 ciclos, cerca de 4 meses, a partir da probabilidade deste consultor ser ativo em cada um dos períodos. Sendo possível identificar grupos de consultores que poderão receber investimentos distintos, de acordo com a faixa de lucratividade de cada grupo.

Résumé : Le travail de recherche présenté dans ce mémoire fut inspiré par le processus de la photosynthèse qui se produit chez les plantes. Au cours de ce processus l’énergie solaire est convertie en énergie chimique via différentes étapes de transferts d’électrons et d’énergie. En maîtrisant bien ces concepts, de nombreuses applications, telles que les cellules photovoltaïques ou les DEL (Diodes électro-luminescentes) peuvent être améliorées. Pour se faire, il est important d’optimiser les propriétés des matériaux existants (oligomères, polymères, etc…) en préparant des systèmes conjugués plus efficaces, mais aussi de pleinement comprendre les processus qui s’y produisent (processus de transferts d’électrons et d’énergie photo-induist). La série d’oligomères et de polymères présentée dans ce mémoire le sont pour leurs applications dans des systèmes photoniques. Dans cette optique, ce mémoire a été divisé en cinq grands chapitres. Le premier présente les principes théoriques de la photophysique. Le second présente le suivi du transfert d’énergie T[indice inférieur 1] dans les états triplets, T[indice inférieur 1], une dyade constituée de la tétraphénylporphyrine de zinc(II), [ZnTPP], et de la bis(phénylpyridinato)(bipyridine) d’iridium(III), [Ir], chromophores liés avec un pont trans-diéthynylbis(phosphine)-platine(II). Malgré que cette dyade soit entièrement conjuguée et qu’elle soit constituée d’un donneur ([ZnTPP]) et d’un accepteur ([Ir]), aucun transfert d’énergie T[indice inférieur 1] [Ir] → S[indice inférieur 1]/T[indice inférieur 1] [ZnTTP] n’a été observé. Ce résultat fut attribué à l’absence de recouvrement des orbitales moléculaires entre la HSOMO(donneur*) et la HSOMO(accepteur), LSOMO(accepteur) and LSOMO (donneur*) (mécanisme de Dexter). Ainsi, l’échange d’électrons est impossible. Ce chapitre suggère que l’équation de Dexter, k[indice inférieur Dexter] = KJexp(-2r[indice inférieur DA]/L) ne reste qu’une approximation. Ce travail a été publié dans ChemComm (2013, 49, 5544-5546). Le troisième chapitre présente le transfert d’énergie singulet beaucoup lent qu’attendu se produisant dans une dyade constituée d’une porphyrine de zinc(II) avec une porphyrine base libre liées par un pont palladium(II) (trans-PdI[indice inférieur 2]). Sachant que cette dyade est entièrement conjuguée et que la distance entre les deux centres de masse des porphyrines est relativement courte, ce système aurait dû présenter un transfert d’énergie très rapide, d’après la théorie de Förster. Dans ce cas, ce comportement a été expliqué par le faible recouvrement des orbitales frontières (OM) du donneur et de l’accepteur. Ce travail a été accepté le 2014-05-26 dans Chemistry – A European Journal (chem.201403146). Le quatrième chapitre rapporte une étude du transfert d’énergie ultra-rapide (650 fs) entre des états singulets dans une dyade composé d’une porphyrine de zinc(II) (le donneur) et une porphyrine base libre (l’accepteur) liées à l’aide d’un pont de palladium ([beta],[beta]--trans-Pd(NH)[indice inférieur 2](CO)[indice inférieur 2]). Ces résultats ont été attribués à la présence d’un couplage fort entre les OM du donneur et de l’accepteur et de la très faible contribution (atomique) du Pd(II) vers ces OM. Cette dyade montre la plus rapide constante de transfert d’énergie k[indice inférieur ET] que nous connaissons pour des dyades similaires contentant un fragment métallique. Les résultats du troisième et quatrième chapitre montrent que la théorie de Förster tel quel ne suffit pas pour prédire les vitesses de transferts d’énergie dans certains systèmes : d’autres facteurs doivent être pris en compte. Ce travail a été soumis dans JACS ( ja-2014-061774, 19-6-2014). Dans le cinquième chapitre de ce mémoire, la synthèse du bis(-[alpha]-(amino(4-éthynylbenzene (triméthylsilane)(R))))bis(4-éthynylbenzene-(triméthylsilane))quinone diimine (R= H, Boc) comme modèle pour des polymères conjugués et non-conjugués contenant le colorant porphyrine a été proposée. Le corps du composé désiré (tétrakis(4-éthynlyphenyl)quinone-1,4-diimine-2,5-diamine) a montré un transfert de charge partant des groupes terminaux riches en électrons une la benzoquinone centrale plus pauvre. La nature de l’émission fut observée uniquement à 77K pour le cas où R = H et fut attribuée à de la fluorescence. À température ambiante, l’intensité était trop faible pour être observée. Dans le cas où R = Boc, aucune emission n’a été détectée. Malheureusement, le composé espéré ne fut pas obtenu, le procédé de synthèse employé engendra uniquement la forme réduite. Cette forme fut malgré tous analysé, et ne présenta pas de transfert de charge ni de communication entre les différents chromophores. Ceci a été expliqué simplement par le fait que la conjugaison est brisée quand ce composé est sous sa forme réduite. Ce travail sera soumis au Journal of Inorganic and Organometallic Polymers and Materials. // Abstract : The research work presented in this master thesis is inspired by the photosynthetic process occurring in plants where solar energy is converted into chemical energy via several energy and electron transfer processes. In the light of these concepts, several applications such as solar cells and light emitting diodes can be improved. To do so, we need to optimize the properties of polyads, oligomers and polymers to device more efficient conjugated materials as well as developing a full understanding of the photo-induced energy and electron transfer processes that occur. Several organometallic oligomers and polymers are presented in this thesis due to their potential photonic applications. In this respect, this master thesis has five chapters. The first one introduces some theoritical principles of photophysics. The second one presents the monitoring of triplet state (T[subscript 1]) energy transfer in a dyad that consists of zinc(II)tetraphenylporphyrin, [ZnTPP], and bis(phenylpyridinato)-(bipyridine)iridium(III), [Ir], chromophores linked by a platinum(II) containing bridge. Despite the conjugation in this dyad and the presence of the [ZnTPP] energy donor and the [Ir] energy acceptor species, no T[subscript 1] [Ir] → S[subscript 1]/T[subscript 1] [ZnTTP] energy transfer occurs. This result was explained by the absence of MO overlap between HSOMO(donor*) and HSOMO(acceptor), LSOMO(donor*) and LSOMO(acceptor) , and hence no efficient double electron transfer exchange (i.e. Dexter mechanism) is likely to occur. This chapter suggested that Dexter formulation, k[subscript Dexter] = KJexp(-2r[subscript DA]/L), appears as an approximation. This work has been published in ChemComm (2013, 49, 5544-5546). The third chapter shows an unexpected slow singlet energy transfer in a dyad built upon a zinc(II)porphyrin and the corresponding free base chromophores linked by a palladium(II)- containing bridge (trans-PdI[subscript 2]), despite the presence of conjugation and the relative short center-to-center distance. This behavior was explained by two factors, the first is the lack of large molecular orbitals (MOs) overlaps between the frontier MOs of the donor and acceptor, and thus preventing a double electron exchange to occur through the trans-PdI[subscript 2] bridge. The second factor affected the energy transfer is the electronic shielding induced by the presence of this same linker, namely the electron rich iodides, preventing the two VI chromophores to fully interact via their transition dipoles. This work has been accepted on 2014-05-26 in Chemistry-A European Journal (chem.201403146). The fourth chapter reports an ultrafast singlet energy transfer (650 fs) in a dyad composed of a zinc(II)porphyrin (donor) and a free base porphyrin (acceptor) [beta],[beta]-linked via trans- Pd(NH)[subscript2](C=O)[subscript 2]. These results were explained by the presence of strong MO couplings of the donor and acceptor and the very weak atomic contribution of the Pd(II) atom to this MO. This dyad shows the fastest energy transfer rate k[subscript ET] among other similar dyad systems incorporating a bridge either in the form of a metal fragment or carbon-based. The results of these third and fourth chapters showed that the Förster mechanism is not enough to account for the energy transfer in some systems and other factors affect that transfer. This work has been submitted in JACS ( ja-2014-061774, 19-6-2014). In chapter 5, the synthesis of bis-[alpha]-(amino(4-ethynylbenzene (trimethylsilane)(R))bis(4- ethynylbenzene-(trimethylsilane))quinone diimine (R = H, Boc) as a model for conjugated and unconjugated porphyrin dye polymers was proposed. The central core of the desired compound, tetrakis(4-ethynlypenyl)quinone-1,4-diimine-2,5-diamine, provided evidence for a charge transfer interaction from the electron richer terminal groups to be more electron poorer benzoquinone ring. The nature of the emission of the core compound was found to be fluorescence at 77K for the case R = H but was too weak to be observed at 298K. No emission was detected for the case R = Boc. Unfortunately, the synthetic route of the desired compound gave the reduced form. The analyses of the reduced compound showed the complete absence of the charge transfer or any communication between the different chromophores due to the broken conjugation between the porphyrin units in the reduced product. This work will be submitted to Journal of Inorganic and Organometallic Polymers and Materials.

Customer lifetime value models (CLTV) are a critical component of customer relationship management strategies. Over time, numerous approaches have been used to estimate the lifetime value (LTV) of a customer or segment of customers to make appropriate decisions on how to distribute marketing dollars and make other customer- related business decisions. In recent years, the development of lower cost data warehousing strategies and the ease with which customer data is captured has increased the volume of data available to firms to utilize in such models. This is, in part, a result of the rise in use of the Internet to interact with customers. Even with the additional data available from Internet interactions, much of the current research in this field relies on membership, subscription based, or contract term data, with little, if any research addressing today's multi-channel retail environment. The robustness of data available for use in application to customer lifetime value models is another result coming from the combination of increased volume of data available, along with advances in the fields of data warehousing and data mining techniques. Existing statistical models for predicting LTV have limitations. Recent advances in machine learning techniques have allowed researchers to apply these techniques to problems similar to customer lifetime value estimation. These techniques can be applied to LTV models. This dissertation develops and evaluates methods for estimating LTV in a multi-channel retail environment. It builds on existing models and introduces supervised learning methods, specifically feed-forward neural networks and regression trees into the prediction models to develop and evaluate new methods for LTV modeling in multi-channel retail environments. The new models proposed by this dissertation present an easier-to-implement solution to predicting churn and the future purchase value of a customer, which are the two key elements of LTV models. These elements provide the multi-channel retail firm with data comparable in customer relationship management utility to LTV data used by organizations whose customer value is rooted in membership, subscription, or contract term data.