A Previsão de Safras tem se constituído em importante ferramenta para o estabelecimento de políticas agrícolas públicas e privadas. Em geral, a Previsão de Safras consiste na previsão do clima e na estimativa do rendimento das partes de interesse econômico de uma cultura. A previsão do clima pode ser realizada pela análise de séries históricas dos parâmetros climáticos e dos efeitos de fenômenos conhecidos, a exemplo do El Niño Oscilação Sul (ENSO), o qual pode ser medido pelo Índice de Oscilação Sul (IOS). Também pode ser realizada pela integração numérica das equações diferenciais que regem os movimentos da atmosfera no planeta Terra, também conhecida como previsão numérica. A estimativa do rendimento das culturas também pode ser realizada pela análise estatística de séries históricas ou pela integração numérica de equações diferenciais que regem a fisiologia e o desenvolvimento das plantas, ambos conhecidos como modelo de culturas. O principal objetivo deste trabalho foi propor uma metodologia para a Previsão de Safras no Brasil, tendo como ponto de partida e protótipo o estudo do rendimento de grãos de milho na região Centro-Sul do país. Para tanto, séries históricas com 60 anos de precipitação pluvial em 24 locais da região Centro-Sul do Brasil foram comparadas aos Índices de Oscilação Sul medidos no mesmo período, inferindo-se que o fenômeno ENSO apresenta efeito marcante, e distinto, apenas em locais mais ao Sul e a Nordeste da região Centro-Sul. Concluiu-se pela impossibilidade de utilização do IOS para a previsão de parâmetros climáticos diários, o que também é prejudicado pela carência de séries históricas dos parâmetros climáticos com 60 ou mais anos no Brasil. Ainda quanto à previsão do clima, as previsões de radiação solar, precipitação pluvial, temperaturas máxima e mínima e umidade relativa do ar, geradas pelo modelo Eta a cada seis horas entre os dias 16/7/1997 e 15/6/2002, foram comparadas às respectivas medidas diárias desses parâmetros climáticos, concluindo-se pela possibilidade da aplicação das previsões geradas pelo modelo Eta na Previsão de Safras, à exceção dos locais mais ao Sul e mais a Nordeste da região Centro-Sul do Brasil. Acerca da estimativa do rendimento de grãos de milho, foi proposto um modelo de cultura baseado na integração das equações que regem a fisiologia e o desenvolvimento das plantas. Comparando-se os rendimentos de grãos de milho estimados nos 24 locais durante as safras 1997/98 a 2001/02, conclui-se pela possibilidade da estimativa do rendimento de grãos de milho na região Centro-Sul pelo modelo proposto. Porém, as discrepâncias entre os rendimentos estimados e os respectivos rendimentos medidos nos locais mais ao Sul e nos locais com textura de solo arenosa apontam a necessidade de correção da estimativa da dinâmica de água realizada pelo modelo de cultura proposto. Como conclusão geral, verificou-se que a metodologia proposta para a Previsão de Safras tem virtudes que devem ser exploradas no sentido de sua implementação no Brasil. Porém, essa implementação depende substancialmente da gestão dos trabalhos, de modo a propiciar as condições necessárias. Cabe destacar que o país tem realizado notáveis avanços nesse setor, caso da implementação da rede meteorológica nacional e do conhecimento gerado pelo Centro de Estudos e Previsões do Clima e pela Empresa Brasileira de Pesquisa Agropecuária, entre outras instituições. Ainda assim, essa área do conhecimento, fundamental para um país agrícola como o Brasil, carece de estudos. / Crop forecast has become an important tool for the private and public agricultural policies to be established. Generally, crop forecast is composed by climatic forecast and the yield estimative of growth of economically interesting parts of crops. The climatic forecast can be performed through the analyses of historical series of the climatic features and of the known phenomena, such as the El Niño Southern Oscillation (ENSO), which can be measured by the Southern Oscillation Index (IOS). It can also be done through a numerical integration of differential equations that rule the atmospheric movements of the Earth, a.k.a. numerical forecast. The estimate of crop yields can also be done through the statistical analysis of historical series or through the integration of differential equations that rule the plant physiology and development, both known as crop models. The main objective of this study was to indicate a methodology for Crop Forecast in Brazil, having as a starting point and prototype the study of corn grain yield in the Center-South region of Brazil. Thus, historical series of 60 years of precipitation in 24 sites of the studied region were compared to the IOS measured in the same period, inferring that the phenomenon ENSO has a remarkable effect, distinctly in the most southern and northeast portions of the studied region. One concluded due to the impossibility of using the IOS for daily climatic forecast, which is threatened by the lack of historical series of climatic features with 60 or more years in Brazil. Regarding the climatic forecast, the forecasts of solar radiation maximum and minimum temperatures and air moisture generated by the model Eta on every 6 hours between July 16, 1997 and June 15, 2002 were compared to the respective daily measurements of these climatic parameters. This provided subsidies for the conclusion that the data generated by the model Eta could be used in the Crop Forecast, except for the most southern and northeast regions in the Center-South region of Brazil. For the estimate of corn grain yield, a model based in the integration of equations that rule the plant physiology and development was proposed. Comparing corn grain yields estimated in 24 sites from the agricultural year 1997/98 to 2001/02, one concluded the possibility of estimating the corn grain yield for the studied region by the proposed model. Although the differences between the estimated and the measured yields in the most southern sites and in those with sandy soils indicate the demand for correction of the estimative of water dynamics performed by the proposed model. As a general conclusion, the methodology proposed for crop forecasting brings positive points which should be explored in the sense of its implementation in Brazil. On the other hand, this implementation depends substantially on the work management, propitiating the necessary conditions. One should highlight that the country has developed notably in this sector, such as the cases of the implementation of the national meteorological net and of the knowledge broadcasted by the Center of Climatic Studies and Forecasting and by the The Brazilian Agricultural Research Corporation (EMBRAPA), among other institutions. Even though, this area of knowledge - vital to an agricultural country as Brazil - demands more research.

O crescimento populacional e a degradação dos corpos d’água vêm exercendo pressão à agricultura moderna, a proporcionar respostas mais eficientes quanto ao uso racional da água. Para uma melhor utilização dos recursos hídricos, faz-se necessário compreender o movimento da água na natureza, onde o conhecimento prévio dos fenômenos atmosféricos constitui uma importante ferramenta no planejamento de atividades que utilizam os recursos hídricos como fonte primária de abastecimento. Nesse trabalho foram realizadas previsões de longo prazo com antecedência de sete meses e intervalo de tempo mensal de níveis no Sistema Hidrológico do Taim, utilizando previsões de precipitação geradas por um modelo de circulação global. Para realizar as previsões foi elaborado um modelo hidrológico empírico de regressão, onde foram utilizadas técnicas estatísticas de análise e manipulação de séries históricas para correlacionar os dados disponíveis aos níveis (volumes) de água no banhado. Partindo do pressuposto que as previsões meteorológicas são a maior fonte de incerteza na previsão hidrológica, foi utilizada a técnica de previsão por conjunto (ensemble) e dados do modelo COLA, com 30 membros, para quantificar as incertezas envolvidas. Foi elaborado um algoritmo para gerar todas as possibilidades de regressão linear múltipla com os dados disponíveis, onde oito equações candidatas foram selecionadas para realizar as previsões. Numa análise preliminar dos dados de entrada de precipitações previstas foi observado que o modelo de circulação global não representou os extremos observados de forma satisfatória, sendo executado um processo de remoção do viés. O modelo de empírico de simulação foi posteriormente executado em modo continuo, gerando previsões de longo prazo de níveis para os próximos sete meses, para cada mês no período de junho/2004 a dezembro/2011. Os resultados obtidos mostraram que a metodologia utilizada obteve bons resultados, com desempenho satisfatórios até o terceiro mês, decaindo seu desempenho nos meses posteriores, mas configurando-se em uma ferramenta para auxílio à gestão dos recursos hídricos do local de estudo. / Population growth and degradation of water bodies have been pressuring modern agriculture, to provide more efficient responses about the rational use of water. For a better use of water resources, it is necessary to understand the movement of water in nature, where prior knowledge of atmospheric phenomena is an important tool in planning activities that use water as the primary source of supply. In this study were performed long-term forecasts of water levels (seven months of horizon, monthly time-step) in the Hydrological System Taim, using rainfall forecasts generated by a global circulation model as input. To perform predictions was developed an empirical hydrological regression model. This model was developed based on statistical techniques of analysis and manipulation of historical data to correlate the input data available to the levels (volume) of water in a wetland. Assuming that weather forecasts are a major source of uncertainty in hydrological forecasting, we used an ensemble forecast from COLA 2.2 with 30 members to quantify the uncertainties involved. An algorithm was developed to generate all the multiple linear regression models with the available data, where eight candidates equations were selected for hydrological forecasting. In a preliminary analysis of the precipitation forecast was observed that the global circulation model did not achieve a good representation of extremes values, thus a process of bias removal was carried out. Then the empirical model was used to generate water levels forecast for the next seven months, in each month of the period june/2004 to december/2011. The results showed that the methodology used has a satisfactory performance until the lead time three (third month in the future) where the performance starts to show lower values. Beside the sharply lost of performance in the last lead times, the model is a support tool that can help the decision making in the management of water resources for the study case.

Combining multiple forecasts in order to generate a single, more accurate one is a well-known approach. A simple average of forecasts has been found to be robust despite theoretically better approaches, increasing availability in the number of expert forecasts, and improved computational capabilities. The dominance of a simple average is related to the small sample sizes and to the estimation errors associated with more complex methods. We study the role that expert correlation, multiple experts, and their relative forecasting accuracy have on the weight estimation error distribution. The distributions we find are used to identify the conditions when a decision maker can confidently estimate weights versus using a simple average. We also propose an improved expert weighting approach that is less sensitive to covariance estimation error while providing much of the benefit from a covariance optimal weight. These two improvements create a new heuristic for better forecast aggregation that is simple to use. This heuristic appears new to the literature and is shown to perform better than a simple average in a simulation study and by application to economic forecast data.

本論文的目的，在探討財務分析師盈餘預測修正的決策過程。因為財務分析師可以說是市場上的專業分析者，故其決策過程特別讓人感到興趣。藉著相關性的迴歸分析，將讓我們了解到分析師的盈餘預測修正可能受到修正前何種事件的影響。 　　本研究依據文獻探討及延伸發展出四個可能會影響分析師盈餘預測正修正方向及幅度的變數，分別是分析師預測修正期間（兩次預測的間隔期）的累積股票報酬率、未預期到的公司管理當局盈餘預測、未預期到的公司期中盈餘以及未預期到的公司期中銷貨。研究期間是從民國七十九年至八十二年，研究的對象則是「財訊」月刊內的研究分析人員（財務分析師）在月刊上所刊載的上市公司盈餘預測。 　　實證結果顯示，在對分析師預測修正的幅度上，以未預期的公司管理當局盈餘預測及未預期的公司期中盈餘兩者有影響力，且如假說預期般地與因變數呈現正相關。在對分析師預測修正的方向上，則四個自變數皆有影響能力，其中前三個自變數與分析師預測修正方向呈現正相關，但最後一個變數則與假說相反地呈現負相關。 　　本研究結果背後可能的原因探討如下：公司管理當局發佈的盈餘預測和公司期中盈餘兩者對分析師的盈餘預測修正方向與幅度上皆有正向影響力，可能意味著此二種資訊的品質具參考性，故才值得專業分析師的青睞。而分析師預測修正期間的累積股票報酬率只對盈餘預測修正方向上有正向的解釋力，暗示著分析師在預測修正時並未完全重視此項資訊。至於公司的未預期期中銷貨對分析師的盈餘預測修正方向上有負向的影響作用，則此異常現象留待後續研究者進一步的探討與測試。

Hurricane Katrina of 2005 was responsible for at least 81 billion dollars of property damage. In planning for such emergencies, society must decide whether to invest in the ability to evacuate more speedily or in improved forecasting technology to better predict the timing and intensity of the critical event. To address this need, we use dynamic programming and Markov processes to model the interaction between the emergency response system and the emergency forecasting system. Simulating changes in the speed of evacuation and in the accuracy of forecasting allows the determination of an optimal mix of these two investments. The model shows that the evacuation improvement and the forecast improvement give different patterns of impact to their benefit. In addition, it shows that the optimal investment decision changes by the budget and the feasible range of improvement.

The future is uncertain. We therefore make predictions and forecasts of the future in order to be able to plan and react to future events. For this purpose, financial analysts are argued to have a responsibility towards investors and the market, in helping to keep the market efficient. Given that financial analysts act in a rational way we argue that analysts should strive to maximize forecast accuracy. The purpose of this study is to investigate how accurate financial analysts’ forecasts of Swedish firms’ future values are, and what information that analysts use that significantly affect the analysts’ forecast accuracy. To investigate this we first examine whether financial analysts contribute with value to investors by comparing their forecast precision against a simple time-series model. Our findings show that financial analysts produce significantly more accurate forecasts than a time-series model in the short term. Furthermore, given that rational analysts act in their own best interest while making accurate forecasts, we argue that analysts will incorporate and use the information that is available to them for the purpose of maximizing forecast accuracy. We investigate this by testing if the analysts’ forecast accuracy is affected by; the forecast horizon, the number of analysts following a firm, the firm size, the corporate visibility, the predictability of earnings, and trading volume. We find that the forecast accuracy is better when the amount of analysts following a firm is high, the firm size is larger, the forecasted company’s corporate visibility in the news is more frequent, and the predictability of earnings is higher. The trading volume does not have a significant effect on analysts’ forecast accuracy. To conclude, we question the value of financial analysts’ forecasts for longer forecast horizons.

This thesis examines the impact of earnings announcements on the stock return performance. Most literature regarding this topic is related to the US market. We follow 40 of the largest and most liquid stocks on the virtual OMX Nordic Exchange from 2010 to 2012. In this research paper, we present the theoretical framework that gives an overview of the possible research areas, and provide empirical evidence of the repercussion of the earnings announcements on stock returns. We use the event study methodology to conduct this thesis. It is a standard approach established by Fama et al. (1969). It has been used in a variety of researches for gauging the effect of new information on the market value of a security. As we expected good news and bad news to have different reactions on the stock return performances, we have split our data in good news and bad news. To differentiate good news from bad news, we measure analysts’ forecast error. It consists in subtracting the earnings per share (EPS) of the analysts’ consensus forecast from the reported EPS of the same year. The analysis is composed of three different subdivisions: the study of the abnormal return during an event window of 17 days, the cumulative abnormal return during this event window, stock price behavior from growth stocks and from value stocks. Our findings show that stock behavior gradually responds to the earnings announcement. The stock reactions that appear within pre-event window may indicate information leakage. Our results describe most average abnormal returns as statistically insignificant during the event window. Earnings information has a lower impact on the stock market. We also find that the effect of positive earnings surprise on stock price lasts longer than that of negative earnings surprise. Stocks from OMX Nordic 40 index have a stable reaction on negative earnings surprise. As a conclusion, we highlight three points. Earning interim and annual earning information disclosure were unable to influence the stock market effectively, and therefore could not fully reflect the changes on the stock price. Investors can get the abnormal returns by using this earnings information during the whole event window.

Starting from 2005 the disclosure of financial forecast for Taiwanese public companies has not been mandatory, firms can decide whether they want to disclose, and if so, how and when to disclose. How does the investor's reaction affect this decision? Furthermore, what is the trade-off between transparency and precision? This study develops a theoretical model in which the voluntary disclosure of earnings forecast is a double-edged sword. Such disclosure may reduce information asymmetry, but simultaneously allows entrepreneurs to hype the stock. The proposed model assumes that insiders might manipulate information and investors can learn with bounded rationality. The analytical results demonstrate that entrepreneurs may forgo earnings forecast disclosure if they can achieve greater profit under non-disclosure. In the multiperiod case, this study shows that insiders would reduce their forecast manipulation behavior due to the cost of forecast error and diminishing marginal expected profit. This study accommodates an explanation of the decrease in voluntary disclosure and the popularity of investor conferences in Taiwan. The inferences of the proposed model are examined based on forecasts issued by Taiwanese listed firms. The empirical results evidence a positive relationship between insiders¡¦ trading profit and manipulation of earnings forecast. Additionally, insiders¡¦ trading profit regarding forecast revisions is greater under voluntary disclosure than mandatory disclosure. This study offers important insights into earnings forecast policy in emerging markets.

Hurricane Katrina of 2005 was responsible for at least 81 billion dollars of property damage. In planning for such emergencies, society must decide whether to invest in the ability to evacuate more speedily or in improved forecasting technology to better predict the timing and intensity of the critical event. To address this need, we use dynamic programming and Markov processes to model the interaction between the emergency response system and the emergency forecasting system. Simulating changes in the speed of evacuation and in the accuracy of forecasting allows the determination of an optimal mix of these two investments. The model shows that the evacuation improvement and the forecast improvement give different patterns of impact to their benefit. In addition, it shows that the optimal investment decision changes by the budget and the feasible range of improvement.

none