Thesis (doctoral)--Kath. Universiẗat, Eichstätt, 2003.

The purpose of confidence and prediction intervals is to provide an interval estimation for an unknown distribution parameter or the future value of a phenomenon. In many applications, prior knowledge about the distribution parameter is available, but rarely made use of, unless in a Bayesian framework. This thesis provides exact frequentist confidence intervals of minimal volume exploiting prior information. The scheme is applied to distribution parameters of the binomial and the Poisson distribution. The Bayesian approach to obtain intervals on a distribution parameter in form of credibility intervals is considered, with particular emphasis on the binomial distribution. An application of interval estimation is found in auditing, where two-sided intervals of Stringer type are meant to contain the mean of a zero-inflated population. In the context of time series analysis, covariates are supposed to improve the prediction of future values. Exponential smoothing with covariates as an extension of the popular forecasting method exponential smoothing is considered in this thesis. A double-seasonality version of it is applied to forecast hourly electricity load under the use of meteorological covariates. Different kinds of prediction intervals for exponential smoothing with covariates are formulated. / Konfidenz- und Prognoseintervalle dienen der Intervallschätzung unbekannter Verteilungsparameter und künftiger Werte eines Phänomens. In vielen Anwendungen steht Vorinformation über einen Verteilungsparameter zur Verfügung, doch nur selten wird außerhalb von bayesscher Statistik davon Gebrauch gemacht. In dieser Dissertation werden exakte frequentistische Konfidenzintervalle unter Vorinformation kleinsten Volumens dargelegt. Das Schema wird auf Verteilungsparameter für die Binomial- und die Poissonverteilung angewandt. Der bayessche Ansatz von Intervallen für Verteilungsparameter wird in Form von Vertrauensintervallen behandelt, mit Fokus auf die Binomialverteilung. Anwendung findet Intervallschätzung in der Wirtschaftsprüfung, wo zweiseitige Intervalle vom Stringer-Typ den Mittelwert in Grundgesamtheiten mit vielen Nullern enthalten sollen. Im Zusammenhang mit Zeitreihenanalyse dienen Kovariaten der Verbesserung von Vorhersagen zukünftiger Werte. Diese Arbeit beschäftigt sich mit exponentieller Glättung mit Kovariaten als eine Erweiterung der gängigen Prognosemethode der exponentiellen Glättung. Eine Version des Modells, welche doppelte Saison berücksichtigt, wird in der Prognose des stündlichen Elektrizitätsbedarfs unter Zuhilfenahme von meteorologischen Variablen eingesetzt. Verschiedene Arten von Prognoseintervallen für exponentielle Glättung mit Kovariaten werden beschrieben.

I denna rapport presenteras ett tillvägagångssätt för att hitta och datera brytpunkter i tidsserier. En brytpunkt definieras av det datum då det skett en stor nivåförändring i tidsserien. Det presenteras även en strategi för att skatta effekten av daterade brytpunkter. Genom att analysera tidsserier över AFA Försäkrings ärendeinflöde visar det sig att brytpunkter i tidsserien sammanfaller med exogena händelser som kan ha orsakat dessa brytpunkter, till exempel villkors- eller lagförändringar inom försäkringsbranschen. Rapporten visar att det genom ett metodiskt angreppssätt går att skatta effekten av en exogen händelse. Dessa skattade effekter kan användas vid framtida prognoser då en liknande förändring förväntas inträffa. Dessutom skapas prognoser över ärendeinflödet två år framåt med olika tidsseriemodeller.

The aim of this study is to get some approach which can help in improving the predictions of daily temperature of Umeå. Weather forecasts are available through various sources nowadays. There are various software and methods available for time series forecasting. Our aim is to investigate the daily maximum temperatures of Umeå, and compare the performance of some methods in forecasting these temperatures. Here we analyse the data of daily maximum temperatures and find the predictions for some local period using methods of autoregressive integrated moving average (ARIMA), exponential smoothing (ETS), and cubic splines.  The forecast package in R is used for this purpose and automatic forecasting methods available in the package are applied for modelling with ARIMA, ETS, and cubic splines. The thesis begins with some initial modelling on univariate time series of daily maximum temperatures. The data of daily maximum temperatures of Umeå from 2008 to 2013 are used to compare the methods using various lengths of training period. On the basis of accuracy measures we try to choose the best method. Keeping in mind the fact that there are various factors which can cause the variability in daily temperature, we try to improve the forecasts in the next part of thesis by using multivariate time series forecasting method on the time series of maximum temperatures together with some other variables. Vector auto regressive (VAR) model from the vars package in R is used to analyse the multivariate time series. Results: ARIMA is selected as the best method in comparison with ETS and cubic smoothing splines to forecast one-step-ahead daily maximum temperature of Umeå, with the training period of one year. It is observed that ARIMA also provides better forecasts of daily temperatures for the next two or three days. On the basis of this study, VAR (for multivariate time series) does not help to improve the forecasts significantly. The proposed ARIMA with one year training period is compatible with the forecasts of daily maximum temperature of Umeå obtained from Swedish Meteorological and Hydrological Institute (SMHI).

O estudo comparou o desempenho preditivo dos modelos de previsão de redes neurais e de suavização exponencial, empregando dados diários do preço da arroba do boi gordo futuro (BM&FBOVESPA) entre janeiro de 2010 até dezembro de 2015. Os resultados mostram que modelos relativamente mais complexos como redes neurais não necessariamente apresentam melhor desempenho se comparados a modelos mais simples, e também mostram que a classificação relativa muda conforme variam as medidas de ajuste e/ou horizonte de previsão além de vantagens associadas a combinação de diversos modelos. / This study compared the predictive performance between neural network models and exponential smoothing, using daily data of live cattle future price (BM&FBOVESPA) from January 2010 to December 2015. The results show that relatively more complex models like neural networks do not necessarily display better performance compared to simpler ones. Results also shows that relative classification changes with respect to adjust measures and/or forecast horizons changes besides advantages achieved by model combinaion.

O estudo comparou o desempenho preditivo dos modelos de previsão de redes neurais e de suavização exponencial, empregando dados diários do preço da arroba do boi gordo futuro (BM&FBOVESPA) entre janeiro de 2010 até dezembro de 2015. Os resultados mostram que modelos relativamente mais complexos como redes neurais não necessariamente apresentam melhor desempenho se comparados a modelos mais simples, e também mostram que a classificação relativa muda conforme variam as medidas de ajuste e/ou horizonte de previsão além de vantagens associadas a combinação de diversos modelos. / This study compared the predictive performance between neural network models and exponential smoothing, using daily data of live cattle future price (BM&FBOVESPA) from January 2010 to December 2015. The results show that relatively more complex models like neural networks do not necessarily display better performance compared to simpler ones. Results also shows that relative classification changes with respect to adjust measures and/or forecast horizons changes besides advantages achieved by model combinaion.

Economic stringencies have compelled Canadian hospitals to examine their administrative effectiveness critically. Improved supplies and inventory procedures adopted by leading industrial corporations, suggest that hospitals might benefit from such systems. Lack of the profit incentive, and the high ratio of wages to total expenses in hospitals, have delayed adoption of modern inventory management techniques. This study examined the economic status of Canadian hospitals, and endeavoured to discover whether a computer-based inventory management system, incorporating short-term statistical demand forecasting, would be feasible and advantageous. Scientific forecasting for inventory management is not used by hospitals. The writer considered which technique would be most suited to their needs, taking account of benefits claimed by industrial users. Samples of demand data were subjected to a variety of simple forecasting methods, including moving averages, exponentially smoothed averages and the Box-Jenkins method. Comparisons were made in terms of relative size of forecast errors; ease of data maintenance, and demands upon hospital clerical staffs. The computer system: BRUFICH facilitated scrutiny of the effect of each technique upon major components of the system. It is concluded that either of two methods would be appropriate: moving averages and double exponential smoothing. The latter, when combined with adaptive control through tracking signals, is easily incorporated within the total inventory system. It requires only a short run of data, tracks trend satisfactorily, and demands little operator intervention. The original system designed by this writer was adopted by the Hospital for Sick Children, Toronto, and has significantly improved their inventory management.

[pt] O presente trabalho desenvolve um método híbrido com a finalidade de criar um índice de qualidade para distribuidoras de energia elétrica. Esse método é construído através da fusão do Método de Análise Hierárquica (AHP) e Técnicas de Amortecimento Exponencial. Com isso, é possível avaliar uma distribuidora levando em conta múltiplos critérios e seus diversos índices passados. / [en] This work develops a hybrid method in order to create a quality index for electric power distribution companies. This method is built through the merger of the Analytical Hierarchy Process (AHP) and exponential smoothing techniques. Thus, it is possible to evaluate a distribution company taking into account multiple criteria and its several indexes in the past.

[pt] Esta dissertação teve como objetivo o desenvolvimento e a implantação de uma metodologia de previsão de vendas e dimensionamento de lotes de encomenda num distribuidor atacadista de produtos químicos. Para tanto, abordou técnicas quantitativas de previsão de demanda de curto prazo e medidas de variância dos erros de previsão a fim de suportar decisões empresariais na aplicação da metodologia, capazes de projetar padrões passados num cenário futuro. A aplicação da metodologia possibilitará à empresa a formalização de um processo atualmente subjetivo, outorgando maior precisão na previsão de vendas, redução de custos com estoque e uma base mais concreta para alocação de recursos financeiros. / [en] This thesis has as objective the developing and implantation of a methodology for forecasting sales and design of batch ordering in a wholesale distributor of chemical products. For this purpose, it approached short term quantitative techniques of demand forecast and measures of variance of forecast errors in order to support business decisions on the application of the methodology, able to design past patterns on a future scenario. The application of the methodology will enable the company the formalization of a process currently subjective, granting a greater accuracy in forecasting sales, reduction in the inventory costs and a more concrete basis for resource allocation.

The objective of this paper is to retrospectively evaluate forecast models for polling data, to be used prospectively for the Swedish general election in 2022. One of the simplest ways of forecasting an election result is through opinion polls, and using the latest observation as the forecast. This paper considers five different forecasting models on polling data which are evaluated based on different error measures and the results are compared to previous research done on the same topic. The data in this paper consists of time series data of party-preference polls from Statistics Sweden. When forecasting polling data, the naive forecasting model was the most accurate, but forecasting the election in 2018 resulted in the forecast combinations model being the most accurate. Finally, the models are used to make forecasts on the Swedish general election taking place in September of 2022.