6 resultados para Data-driven Methods
em Repositório digital da Fundação Getúlio Vargas - FGV
Resumo:
Nesta dissertação realizou-se um experimento de Monte Carlo para re- velar algumas características das distribuições em amostras finitas dos estimadores Backfitting (B) e de Integração Marginal(MI) para uma regressão aditiva bivariada. Está-se particularmente interessado em fornecer alguma evidência de como os diferentes métodos de seleção da janela hn, tais co- mo os métodos plug-in, impactam as propriedades em pequenas amostras dos estimadores. Está-se interessado, também, em fornecer evidência do comportamento de diferentes estimadores de hn relativamente a seqüência ótima de hn que minimiza uma função perda escolhida. O impacto de ignorar a dependência entre os regressores na estimação da janela é tam- bém investigado. Esta é uma prática comum e deve ter impacto sobre o desempenho dos estimadores. Além disso, não há nenhuma rotina atual- mente disponível nos pacotes estatísticos/econométricos para a estimação de regressões aditivas via os métodos de Backfitting e Integração Marginal. É um dos objetivos a criação de rotinas em Gauss para a implementação prática destes estimadores. Por fim, diferentemente do que ocorre atual- mente, quando a utilização dos estimadores-B e MI é feita de maneira completamente ad-hoc, há o objetivo de fornecer a usuários informação que permita uma escolha mais objetiva de qual estimador usar quando se está trabalhando com uma amostra finita.
Resumo:
A criação e popularização da Internet têm provocado sérias mudanças no campo da comunicação. Especificamente na comunicação científica, o uso do meio eletrônico permite agilizar as etapas do processo editorial, bem como facilitar a criação, disseminação, pesquisa e uso de informação científica. Entretanto, a incorporação de tecnologias é também desafiadora, pois exige diversas mudanças no modelo de gestão de periódicos científicos. Esse é composto basicamente por aspectos científicos (certificação de conteúdo visando a seleção e divulgação de conhecimento de qualidade), administrativos (produção editorial e gráfica, gestão administrativa e financeira, comunicação e marketing) e financeiros (modelo de rendimentos) que precisam estar alinhados para que o periódico possa cumprir seu objetivo. A presente pesquisa teve como principal objetivo analisar a gestão editorial de periódicos científicos da área de Administração no contexto brasileiro e propor uma tipologia de seus modelos de gestão. Adotou-se de uma abordagem qualitativa e foram utilizados métodos de pesquisa documental, entrevista e análise bibliométrica para, respectivamente, identificação do contexto, descrição dos modelos de gestão editorial e cálculo do impacto dos periódicos selecionados para o estudo (Revista de Administração de Empresas, Revista de Administração Contemporânea, Revista de Administração da USP, Organizações & Sociedade e Brazilian Administration Review). Identificou-se que os periódicos selecionados apresentam impacto (medido por meio de citações) relativamente alto quando comparados aos periódicos estrangeiros avaliados pelo Qualis da área de Administração, Contabilidade e Turismo. Percebe-se que o modelo de gestão dos periódicos está fortemente vinculado ao tipo de Instituição Mantenedora desse (Associação Científica, Instituição de Ensino Superior (IES) Pública, IES Privada), a qual influencia nas questões financeiras, administrativas e científicas das revistas. Espera-se que os resultados da pesquisa possam contribuir não só para os editores de revistas científicas, mas também para os responsáveis por decisões que se referem às políticas de avaliação e fomento de periódicos científicos no Brasil.
Resumo:
We study semiparametric two-step estimators which have the same structure as parametric doubly robust estimators in their second step. The key difference is that we do not impose any parametric restriction on the nuisance functions that are estimated in a first stage, but retain a fully nonparametric model instead. We call these estimators semiparametric doubly robust estimators (SDREs), and show that they possess superior theoretical and practical properties compared to generic semiparametric two-step estimators. In particular, our estimators have substantially smaller first-order bias, allow for a wider range of nonparametric first-stage estimates, rate-optimal choices of smoothing parameters and data-driven estimates thereof, and their stochastic behavior can be well-approximated by classical first-order asymptotics. SDREs exist for a wide range of parameters of interest, particularly in semiparametric missing data and causal inference models. We illustrate our method with a simulation exercise.
Resumo:
This research attempts to analyze the effects of open government data on the administration and practice of the educational process by comparing the contexts of Brazil and England. The findings illustrate two principal dynamics: control and collaboration. In the case of control, or what is called the "data-driven" paradigm, data help advance the cause of political accountability through the disclosure of school performance. In collaboration, or what is referred to as the "data-informed" paradigm, data is intended to support the decision-making process of administrators through dialogical processes with other social actors.
Resumo:
This paper approaches the strategy in business management and aimed at identifying and outlining the interests and commitment of stakeholders in strategic resources management concerning production and implementation of wind turbine equipment of a Brazilian wind power company and also verifying if internal and external results deriving from such activities were sustainable, taking as main reference seminal publications and periodicals relevant to the research point that discuss the Resource Theory, Stakeholders and Sustainability. An analysis was carried out to assess how stakeholders, beyond the temporal context, intermediated the composition, development and management of the organization´s resources, as well as the social, environmental and economic results obtained from resources management in the production and supply of wind turbines to a Wind Power Plant located in the State of Ceara, in order to portray that Brazil sustainability can be an important competitive advantage source that creates value for shareholders and the community (Hart & Milstein, 2003). The strategy herein applied was the qualitative investigation using a single study case, which allowed for the thorough examination of an active organization operating in the Brazilian industry of wind power and also the resources used in the production and implementation of wind turbines supplied to the a Wind Power Plant in Ceara. Considering the content analysis and the triangulation principle, three qualitative data collection methods were applied to identify and characterize stakeholders’ interest and commitment in resource management of the organization operating in the Brazilian wind power industry, as follows, semistructured deep interview with managers of tactic-strategic level and analysts of organization´s value chain nine activities, analysis of public internal and external documents; and analysis of audio-visual material. Nonetheless, to identify the internal and external economic, social and environmental results of implementation and supply of wind turbines to the Wind Power Plant in Ceara, semistructured interviews were also carried out with the residents of the region. Results showed the BNDES (Brazilian Development Bank) and the organization head office were the stakeholders who exerted the strongest influence on resources related to production and implementation of the aerogenerator product at Trairi Wind Plant in Ceara. Concerning the organization resources, at the current stage of the Brazilian Wind Industry ,although the brand, reliability and reputation of the organization under study were valuable esources, rare, hard to imitate and exploited by the organization, it was noticed that opposed to RBV, they did not actually represent a source of competitive advantage . For the local community the social, economic and environmental results related to the wind turbines implementation were more positive than negative, despite the fact that the productive process caused negative environmental impacts such as the high emission of CO2 to transport wind turbines components to Trairi Wind Power Plant.
Resumo:
The synthetic control (SC) method has been recently proposed as an alternative method to estimate treatment e ects in comparative case studies. Abadie et al. [2010] and Abadie et al. [2015] argue that one of the advantages of the SC method is that it imposes a data-driven process to select the comparison units, providing more transparency and less discretionary power to the researcher. However, an important limitation of the SC method is that it does not provide clear guidance on the choice of predictor variables used to estimate the SC weights. We show that such lack of speci c guidances provides signi cant opportunities for the researcher to search for speci cations with statistically signi cant results, undermining one of the main advantages of the method. Considering six alternative speci cations commonly used in SC applications, we calculate in Monte Carlo simulations the probability of nding a statistically signi cant result at 5% in at least one speci cation. We nd that this probability can be as high as 13% (23% for a 10% signi cance test) when there are 12 pre-intervention periods and decay slowly with the number of pre-intervention periods. With 230 pre-intervention periods, this probability is still around 10% (18% for a 10% signi cance test). We show that the speci cation that uses the average pre-treatment outcome values to estimate the weights performed particularly bad in our simulations. However, the speci cation-searching problem remains relevant even when we do not consider this speci cation. We also show that this speci cation-searching problem is relevant in simulations with real datasets looking at placebo interventions in the Current Population Survey (CPS). In order to mitigate this problem, we propose a criterion to select among SC di erent speci cations based on the prediction error of each speci cations in placebo estimations