Thesis (Ph.D.)--Kent State University, 2007. / Title from PDF t.p. (viewed Nov. 14, 2007). Advisor: Ran Barniv. Includes bibliographical references (p. 117-122).

Thesis (Ph. D.)--West Virginia University, 2002. / Title from document title page. Document formatted into pages; contains viii, 175 p. : ill. Includes abstract. Includes bibliographical references (p. 171-175).

Thesis (M.S.)--Rochester Institute of Technology, 2005. / Typescript. Includes bibliographical references (leaves 81-82).

Thesis (M.Eng. (Hons.)) -- University of Western Sydney, 2008. / A thesis submitted towards the degree of Master of Engineering (Honours) in the University of Western Sydney, College of Health and Science, School of Engineering. Includes bibliographical references.

Thesis (Ph. D.)--University of California, San Diego, 2006. / Title from first page of PDF file (viewed September 20, 2006). Available via ProQuest Digital Dissertations. Vita. Includes bibliographical references (p. 125-132).

We will compare two different forecasting models with the forecasting model that was used in March 2014 by The Swedish Social Insurance Agency ("Försäkringskassan" in Swedish or "FK") in this degree project. The models are used for forecasting the number of cases. The two models that will be compared with the model used by FK are the Seasonal Exponential Smoothing model (SES) and Auto-Regressive Integrated Moving Average (ARIMA) model. The models will be used to predict case volumes for two types of benefits: General Child Allowance “Barnbidrag” or (BB_ABB), and Pregnancy Benefit “Graviditetspenning” (GP_ANS). The results compare the forecast errors at the short time horizon (22) months and at the long-time horizon (70) months for the different types of models. Forecast error is the difference between the actual and the forecast value of case numbers received every month. The ARIMA model used in this degree project for GP_ANS had forecast errors on short and long horizons that are lower than the forecasting model that was used by FK in March 2014. However, the absolute forecast error is lower in the actual used model than in the ARIMA and SES models for pregnancy benefit cases. The results also show that for BB_ABB the forecast errors were large in all models, but it was the lowest in the actual used model (even the absolute forecast error). This shows that random error due to laws, rules, and community changes is almost impossible to predict. Therefore, it is not feasible to predict the time series with tested models in the long-term. However, that mainly depends on what FK considers as accepted forecast errors and how those forecasts will be used. It is important to mention that the implementation of ARIMA differs across different software. The best model in the used software in this degree project SAS (Statistical Analysis System) is not necessarily the best in other software.

This dissertation comprises of three original contributions to empirical forecasting research. Chapter 1 introduces the dissertation. Chapter 2 contributes to the literature on hierarchical time series (HTS) modelling by proposing a disaggregated forecasting system for both inflation rate and its volatility. Using monthly data that underlies the Retail Prices Index for the UK, we analyse the dynamics of the inflation process. We examine patterns in the time-varying covariation among product-level inflation rates that aggregate up to industry-level inflation rates that in turn aggregate up to the overall inflation rate. The aggregate inflation volatility closely tracks the time path of this covariation, which is seen to be driven primarily by the variances of common shocks shared by all products, and by the covariances between idiosyncratic product-level shocks. We formulate a forecasting system that comprises of models for mean inflation rate and its variance, and exploit the index structure of the aggregate inflation rate using the HTS framework. Using a dynamic model selection approach to forecasting, we obtain forecasts that are between 9 and 155 % more accurate than a SARIMA-GARCH(1,1) for the aggregate inflation volatility. Chapter 3 is on improving forecasts using forecast combinations. The paper documents the software implementation of the open source R package for forecast combination that we coded and published on the official R package depository, CRAN. The GeomComb package is the only R package that covers a wide range of different popular forecast combination methods. We implement techniques from 3 broad categories: (a) simple non-parametric methods, (b) regression-based methods, and (c) geometric (eigenvector) methods, allowing for static or dynamic estimation of each approach. Using S3 classes/methods in R, the package provides a user-friendly environment for applied forecasting, implementing solutions for typical issues related to forecast combination (multicollinearity, missing values, etc.), criterion-based optimisation for several parametric methods, and post-fit functions to rationalise and visualise estimation results. The package has been listed in the official R Task Views for Time Series Analysis and for Official Statistics. The brief empirical application in the paper illustrates the package’s functionality by estimating forecast combination techniques for monthly UK electricity supply. Chapter 4 introduces HTS forecasting and forecast combination to a healthcare staffing context. A slowdown of healthcare budget growth in the UK that does not keep pace with growth of demand for hospital services made efficient cost planning increasingly crucial for hospitals, in particular for staff which accounts for more than half of hospitals’ expenses. This is facilitated by accurate forecasts of patient census and churn. Using a dataset of more than 3 million observations from a large UK hospital, we show how HTS forecasting can improve forecast accuracy by using information at different levels of the hospital hierarchy (aggregate, emergency/electives, divisions, specialties), compared to the naïve benchmark: the seasonal random walk model applied to the aggregate. We show that forecast combination can improve accuracy even more in some cases, and leads to lower forecast error variance (decreasing forecasting risk). We propose a comprehensive parametric approach to use forecasts in a nurse staffing model that has the aim of minimising cost while satisfying that the care requirements (e.g. nurse hours per patient day thresholds) are met.

Finansinio įmonės modelio tema yra viena iš aktualiausių verslo srityje. Taip yra dėl to, kad informacijos srautai nuolatos didėja, o finansinės ataskaitose pateikiama informacija turi būti susisteminta. Ataskaitos tuo pačiu metu turi suteikti kuo daugiau reikalingos informacijos ir nebūti perkrautos duomenimis. Pateikiamų ataskaitų, sudarytų pagal dvejybinį modelį, duomenų nepakanka, todėl buvo pasiūlytas naujas – trejybinis apskaitos modelis. Darbe yra naudojama mokslinės literatūros analizė, aprašomasis ir palyginimo metodai, statistinė analizė, finansinis modeliavimas ir apibendrinimo metodas. Darbo tikslas yra išnagrinėti ar tikslinga taikyti tolesniam įmonės veiklos prognozavimui dvejybinį apskaitos modelį. Iškelta tyrimo hipotezė, ar trejybinis modelis, su tam tikrom išlygom, yra tinkamesnis finansiniam prognozavimui negu dvejybinis modelis. Kaip parodė galutiniai darbo rezultatai, tikslesnė veiklos prognozė bus galima pritaikius Trejybinį apskaitos modelį, kadangi šis modelis ne tik paaiškina pakeitimus nagrinėjamuosiuose straipsniuose, tačiau jis dar parodo ir veiksnių kitimo dinamiką analizuojamuoju laikotarpiu. Taip pat pateikiamas balansas ir Turto, Momento ir Jėgos ataskaitos, sudarytos trejybinio modelio pagrindu. Parodomi trejybinio modelio pranašumai, suteikiami informacijos gavėjui. Nors trejybinis finansinis modelis palyginus su dvejybiniu yra dar labai jaunas ir iki galo neišnagrinėtas, jį reikėtų išnagrinėti išsamiau, įsigilinant ne tik į patį modelio... [toliau žr. visą tekstą] / Financial modeling is one of the most important theme in business. The reason of this is that the amount of information is growing all the time and the financial statements have to represent information systemically. Statements at the same time have to represnt as much information as it is possible but at the same time the statement shouldn`t be overflouted. The information represented in statements, which are made by double-entry model is not enough for forecasting. That is the reason why a new model for forecasting was proposed. In thesis it is used sientific literature analysis, comparative, systematic and statistical analysis, financial modelling and summative methods. The aim the thesis is to adjust would it be correct to use double-entry bookkeeping for future forecasting or not. Thesis‘s hypethesis supposes that the future forecasting made using triple-entry bookkeeping model, a new one, would be more correct. According to thesis results, forecasting based on triple-entry bookkeeping would be more correct. Besides that, this model shows not only the changes in paper but the dynamics movement during analysed term. Furthemore, in thesis it is based Wealth, Moment and Force statements which was made by triple-entry bookkeeping. The advantages of this model is also put into the text. As it is known, triple-entry bookkeeping is more younger that double-entry and it wasn`t analysed very good yet. Nevertheless, the metodics how triple-entry bookkeeping can be used should be... [to full text]

This thesis deals with the prediction demand forecasting in a company, focusing especially on quantitative methods of prediction. The theoretical part presents the predictions of demand, its place and importance in a company. Secondly, it presents various methods of qualitative and quantitative demand forecasting and the methods for measuring prediction accuracy. The practical part applies several methods on a real data of the company. These are the methods of moving averages, exponential smoothing, Holt and Holt-Winters method and the simple linear regression. The accuracy of each method are compared with each other and most accurate method is then used to predict demand for the year 2015.

Big Data and Big Data analytics have attracted major interest in research and industry and continue to do so. The high demand for capable and scalable analytics in combination with the ever increasing number and volume of application scenarios and data has lead to a large and intransparent landscape full of versions, variants and individual algorithms. As this zoo of methods lacks a systematic way of description, understanding is almost impossible which severely hinders effective application and efficient development of analytic algorithms. To solve this issue we propose our concept of modular analytics that abstracts the essentials of an analytic domain and turns them into a set of universal building blocks. As arbitrary algorithms can be created from the same set of blocks, understanding is eased and development benefits from reusability.