We perform a review of model selection procedures, in particular various cross validation procedures and adaptive model selection. We cover important results for these procedures and explore the connections between different procedures and information criteria.

The work provides the results of the survey on the present situation of salary calculation information systems in the market trying to distinguish advantages and disadvantages of the usable systems, their compatibility with organizations’ administrative information systems as well as possibilities to adjust them to the requirements of users. The survey confirmed that most of examined information systems are either old, or cause many problems then combining them with organizations’ administrative information systems. Moreover, only some of the examined salary calculation information systems provide users with a possibility to change calculation formulas, reports or other properties of information system. This work presents the description of adaptive salary calculation model. This model gives the version of a possible solution of a problem which reduce the influence of determinate salary calculation information systems’ drawbacks. In this work preferred basic diagrams and descriptions of adaptive model of salary calculation. Late was created project of adaptive model of salary calculation according to diagrams and descriptions. Development of adaptive model of salary calculation gives possibility to change calculation formulas, reports and other properties of information system. Analysis gives that adaptive model of salary calculation information system solves the problem and is more comfortable for the user then other salary calculation information systems. According to this... [to full text]

No description available.

<p>Osnovni istraživački problem ove disertacije predstavlja razvoj modela<br />za upravljanje lancima snabdevanja u cilju podizanje kvaliteta usluge.<br />Stoga je razvijen adaptivni model za upravljanje lancima snabdevanja<br />koji se sastoji od modela za: upravljanje lancima snabdevanja,<br />upravljanje korisničkim zahtevima i ocenu kvaliteta pružene usluge. U<br />svrhu primene adaptivni model je predstavljen algoritmom sa precizno<br />definisanim koracima koje korisnik treba da sprovede da bi podigao<br />nivo kvaliteta usluge i održao stabilnost lanca snabdevanja. Verifikacija<br />modela je urađena na primeru 17 lanaca snabdevanja na teritoriji<br />Republike Srbije, &scaron;to je rezultiralo odgovorima na koji način se može<br />podići kvalitet usluge.<br />Doprinos istraživanja ogleda u mogućnosti direktne primene<br />razvijenog modela i pružanja novih informacija za naučnu i stručnu<br />javnost koje mogu predstavljati kvalitetnu podlogu daljem razvoju<br />modela za upravljanje lancima snabdevanja.</p> / <p>The basic research problem of this dissertation is the development of<br />supply chain management model in order to improve the quality of<br />service. Therefore, an adaptive supply chain management model has<br />been developed that consists of a model for: supply chain management,<br />management of user requirements and assessment of the quality of<br />service provided. For the purpose of application the adaptive model, it<br />is presented an algorithm with precisely defined steps that the user<br />needs to implement in order to raise the level of service quality and<br />maintain the stability of supply. The model verification was done on<br />the example of 17 supply chains in the territory of the Republic of<br />Serbia, which resulted in answers on how to improve the quality of the<br />service. The contribution of the research is reflected in the possibility<br />of direct application of the developed model and providing new<br />information for the scientific and professional public, which can<br />represent a quality basis for the further development of the supply<br />chain management model.</p>

Orientadores: Rubens Maciel Filho, Aline Carvalho da Costa / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Quimica / Made available in DSpace on 2018-08-07T06:46:12Z (GMT). No. of bitstreams: 1 Mantovaneli_IvanaCristinaCordeiro_M.pdf: 4744367 bytes, checksum: 648b4b28d0ad3fff1bfeda9851005fc1 (MD5) Previous issue date: 2005 / Resumo: A utilização do etanol como combustível tem muitas vantagens; ele tem competido economicamente com a gasolina e diversos outros combustíveis, substituindo-os em várias utilidades. Desta forma, existe um grande interesse em se otimizar todos os passos da produção de etanol. Apenas um conhecimento profundo da dinâmica do processo gera uma operação ótima, e este pode ser conseguido através de simulações realizadas usando um modelo preciso. Muitos modelos fenomenológicos foram desenvolvidos considerando condições industriais, mas estes só são válidos para condições específicas nas quais foram determinados, invalidando a predição do modelo em outras condições. Mudanças acontecem normalmente em uma unidade industrial e a re-estimação freqüente dos parâmetros do modelo é usualmente difícil e demorada. O objetivo deste trabalho é desenvolver um modelo híbrido neuronal para o processo de fermentação alcoólica usando balanços de massa combinados com redes neuronais do tipo Functional Link. Será implementado um esquema para atualização dos pesos da rede sempre que esta não descrever o comportamento dinâmico da planta. O modelo desenvolvido será usado para descrever um processo real no lugar dos modelos fenomenológicos existentes, já que estes têm sido capazes de descrever o processo apenas por curtos espaços de tempo / Abstract: The use of ethanol as a fuel has many advantages; it has economically competed with gasoline and others fuels, substituting them in various uses. Thus, there is a great interest in optimizing all the steps of ethanol production. Only a detailed knowledge of the process dynamics can lead to optimal operation and this can be achieved through simulation using an accurate model. Many phenomenological models were developed considering industrial fermentations, but they are only valid for specific conditions. Changes occur frequently in an industrial unity and frequent reestimation of model parameters is usually expensive and time consuming. The objective of this work is to develop a hybrid neural model for the alcoholic fermentation process using mass balances combined with Functional Link Neural Networks. A scheme to update network weights always that it does not describe plant behavior accurately is implemented. The developed model is used to describe an industrial process substituting the existing phenomenological models, since they have been able to describe the process only for short periods / Mestrado / Desenvolvimento de Processos Biotecnologicos / Mestre em Engenharia Química

本研究主要分為三個部分：第一個部分探討壽險公司保單組合之最適資產配置；第二個部分探討確定提撥退休金制度下，員工所面臨的資產配置問題；第三個部分則為方法論的比較研究。此外，本文也探討長命風險(longevity risk)等相關議題。本文在Huang與Cairns (2006) 所提出的資產報酬模型下，推導出累積資產價值的期望值以及變異數，並利用套裝軟體的最佳化程式(optimization programming)獲得給定目標函數下的最適投資策略。 在保單組合資產配置之研究方面，我們分別針對保險公司繼續經營的商品以及即將停賣的商品提出合適的資產配置方式。常數資產配置方式(Constant rebalance rule)適合持續經營的商品，變動資產配置方式(Variable rebalance rule)則適合即將停賣的商品。在常數資產配置方式下，我們能夠得到投資組合的效率前緣線。此外，不管是何種資產配置方式，當保單組合的保單到期日較近時，保險公司必須增加其所持有的現金比例。 在確定提撥制下最適資產配置問題的研究方面，本文的結果符合一般退休基金經理人所採取的生命週期型態投資方式。本研究發現在Lee-Carter模型之下，考慮時間加權可以增加模型的預測能力。而在考慮長命風險下，員工必須採取更積極的投資策略。 本文決定資產配置之方法為預期模型(Anticipative model)，其在評價日時即決定未來的決策，不考慮新訊息對決策的影響。考慮新訊息會對決策產生影響的決定資產配置方法為適應模型(Adaptive model)。在第五章的研究裡，我們比較上述兩種決定資產配置方法之差異。研究結果發現，若以期望值與標準差為判斷標準，兩種決定資產配置方法並沒有絕對的優劣關係。而若在每個決策執行的時間點重新使用預期模型來決定新的資產配置策略，則其所對應的投資策略以及投資績效會與適應模型下的策略與投資績效接近。因此，在無法獲得適應模型投資策略封閉解的情況下，預期模型投資策略可以有效的近似適應模型投資策略。 / The purpose of this thesis is to investigate the asset allocation issue of the long-term investors. Our approach is to calculate theoretical formulae of the first two moments of the accumulated fund; we then adopt optimization programming to find a asset allocation strategy that fits the fund management target. Two kinds of investors are explored. The first one is an investment manager who manages a general portfolio of life insurance policies, and the second one is an employee who starts his career life in a DC pension plan. We also survey the longevity risk issue in this thesis. In the study of “optimal asset allocation for a general portfolio of life insurance policies”, two kinds of rebalancing methodologies are examined. For constant rebalance rule, which is applicable to a continuing business line, we find an efficient frontier in the mean-standard deviation plot that occurs with arbitrary policy portfolios. Also, the insurance company should hold more cash to reduce its illiquidity risk for portfolios in which policies will mature at earlier dates. In the study of “optimal asset allocation incorporating longevity risk in defined contribution pension plans”, we confirm the suitability of the lifestyle investment strategy. Investors in a DC pension plan should be more aggressive when he considers the longevity risk. Furthermore, we proposed a time adjustment technique to capture mortality predictions more precisely in this study. The approach of decision making of this thesis is referred to anticipative model, which does not consider the possible feedback from the future information. On the other hand, the approach of decision making that consider the possible feedback from the future information is referred to adaptive model. We further compare the two approached in the study “Comparative efficiency- anticipative model versus adaptive model”. The numerical results show that investors would not prefer the adaptive approach to the anticipative approach in the mean-variance criterion. Moreover, the downside risk is larger when the strategy is decided by adaptive approach. We also find that the strategy and its numerical distribution of anticipative approach can approximate to that of adapted approach if one re-assesses it at every decision date. Thus, the anticipative approach provides a first approximation on looking for the optimal investment strategy of adaptive model.