Over the past few decades, various computational and mathematical methodologies have been proposed for forecasting seasonal epidemics. In recent years, the deadly effects of enormous pandemics such as the H1N1 influenza virus, Ebola, and Zika, have compelled scientists to find new ways to improve the reliability and accuracy of epidemic forecasts. The improvement and variety of these prediction methods are undeniable. Nevertheless, many challenges remain unresolved in the path of forecasting the outbreaks using surveillance data. Obtaining the clean real-time data has always been an obstacle. Moreover, the surveillance data is usually noisy and handling the uncertainty of the observed data is a major issue for forecasting algorithms. Correct modeling assumptions regarding the nature of the infectious disease is another dilemma. Oversimplified models could lead to inaccurate forecasts, whereas more complicated methods require additional computational resources and information. Without those, the model may not be able to converge to a unique optimum solution. Through the last decade, there has been a significant effort towards achieving better epidemic forecasting algorithms. However, the lack of standard, well-defined evaluating metrics impedes a fair judgment on the proposed methods. This dissertation is divided into two parts. In the first part, we present a Bayesian particle filter calibration framework integrated with an agent-based model to forecast the epidemic trend of diseases like flu and Ebola. Our approach uses Bayesian statistics to estimate the underlying disease model parameters given the observed data and handle the uncertainty in the reasoning. An individual-based model with different intervention strategies could result in a large number of unknown parameters that should be properly calibrated. As particle filter could collapse in very large-scale systems (curse-of-dimensionality problem), achieving the optimum solution becomes more challenging. Our proposed particle filter framework utilizes machine learning concepts to restrain the intractable search space. It incorporates a smart analyzer in the state dynamics unit that examines the predicted and observed data using machine learning techniques to guide the direction and amount of perturbation of each parameter in the searching process. The second part of this dissertation focuses on providing standard evaluation measures for evaluating epidemic forecasts. We present an end-to-end framework that introduces epidemiologically relevant features (Epi-features), error measures, and ranking schema as the main modules of the evaluation process. Lastly, we provide the evaluation framework as a software package named Epi-Evaluator and demonstrate the potentials and capabilities of the framework by applying it to the output of different forecasting methods. / PHD / Epidemics impose substantial costs to societies by deteriorating the public health and disrupting economic trends. In recent years, the deadly effects of wide-spread pandemics such as H1N1, Ebola, and Zika, have compelled scientists to find new ways to improve the reliability and accuracy of epidemic forecasts. The reliable prediction of future pandemics and providing efficient intervention plans for health care providers could prevent or control disease propagations. Over the last decade, there has been a significant effort towards achieving better epidemic forecasting algorithms. The mission, however, is far from accomplished. Moreover, there has been no significant leap towards standard, well-defined evaluating metrics and criteria for a fair performance judgment between the proposed methods. This dissertation is divided into two parts. In the first part, we present a Bayesian particle filter calibration framework integrated with an agent-based model to forecast the epidemic trend of diseases like flu and Ebola. We model the disease propagation via a large scale agent-based model that simulates the disease spread across the contact network of people. The contact network consists of millions of nodes and is constructed based on demographic information of individuals achieved from the census data. The agent-based model’s configurations are mostly unknown parameters that should be properly calibrated. We present a Bayesian particle filter calibration approach to estimate the underlying disease model parameters given the observed data and handle the uncertainty in the reasoning. As particle filter could collapse in very large-scale systems, achieving the optimum solution becomes more challenging. Our proposed particle filter framework utilizes machine learning concepts to restrain the intractable search space. It incorporates a smart analyzer unit that examines the predicted and observed data using machine learning techniques to guide the direction and amount of perturbation of each parameter in the searching process. The second part of this dissertation focuses on providing standard evaluation measures for evaluating and comparing epidemic forecasts. We present a framework that introduces epidemiologically relevant features (Epi-features), error measures, and ranking schema as the main modules of the evaluation process. Lastly, we provide the evaluation framework as a software package named Epi-Evaluator and demonstrate the potentials and capabilities of the framework by applying it to the output of different forecasting methods.

One of the fundamental premises of financial reporting by business enterprises is that it should provide users with information that will assist them in predicting the amounts, timing and uncertainty of future cash flows of the enterprise. The requirement for alternative income measurements was partially justified by an assumed correspondence between the new information and the cash flow prediction objective. The existence of that correspondence, however, has not been precisely verified by the research to date. The overall objective of this research was to contribute additional evidence to address conflicts in the prior research findings, and additionally, to consider possible industry and firm-size effects on the ability to predict cash flow from alternative incomes. A data base was compiled from COMPUSTAT tapes (historical cost), the Parker model restatement procedures (general price-level) and the Easman data base that used the Falkenstein-Weil restatement model (replacement cost). One conclusion was that the alternative income measurements produce different cash flow forecast errors. Overall, historical cost net income produced the lowest forecast errors for two approximations of cash flow. The inclusion of monetary gains/losses and holding gains/losses in net income did not improve predictions, and in one case worsened them. Another conclusion was that a multiple linear regression model produced significantly lower forecast errors for both cash flow definitions. The simple linear and exponential regression prediction models did not produce different forecast errors. Finally, both an industry effect and a firm-size effect were identified in the prediction of working capital from operations. When net income plus depreciation was the object of prediction, an industry effect was identified but not a firm-size effect. The overall impact of these findings is that the alternative income measurements should be justified on some basis other than facilitating cash flow prediction. In fact, a random-walk cash flow prediction model performed better than any prediction based on net income. Financial accounting standards in the area of alternative income measurements should consider possible industry and firm-size differences. The choice of cash flow definition is apparently critical because different conclusions were obtained. / Ph. D.

A principal factor contributing to hydrologic prediction uncertainty is modeling error intro- duced by the measurement and prediction of precipitation. The research presented demon- strates the necessity for using probabilistic methods to quantify hydrologic forecast uncer- tainty due to the magnitude of precipitation errors. Significant improvements have been made in precipitation estimation that have lead to greatly improved hydrologic simulations. However, advancements in the prediction of future precipitation have been marginal. This research shows that gains in forecasted precipitation accuracy have not significantly improved hydrologic forecasting accuracy. The use of forecasted precipitation, referred to as quantita- tive precipitation forecast (QPF), in hydrologic forecasting remains commonplace. Non-zero QPF is shown to improve hydrologic forecasts, but QPF duration should be limited to 6 to 12 hours for flood forecasting, particularly for fast responding watersheds. Probabilistic hydrologic forecasting captures hydrologic forecast error introduced by QPF for all forecast durations. However, public acceptance of probabilistic hydrologic forecasts is problematic. Central tendency measures from a probabilistic hydrologic forecast, such as the ensemble median or mean, have the appearance of a single-valued deterministic forecast. The research presented shows that hydrologic ensemble median and mean forecasts of river stage have smaller forecast errors than current operational methods with forecast lead-time beginning at 36-hours for fast response basins. Overall, hydrologic ensemble median and mean forecasts display smaller forecast error than current operational forecasts. / Ph. D. / Flood forecasting is uncertain, in part, because of errors in measuring precipitation and predicting the location and amount of precipitation accumulation in the future. Because of this, the public and other end-users of flood forecasts should understand the uncertainties inherent in forecasts. But, there is reluctance by many to accept forecasts that explicitly convey flood forecast uncertainty, such as, ”there is a 67% chance your house will be flooded”. Instead, most prefer ”your house will not be flooded” or something like ”flood levels will reach 0.5 feet in your house”. We hope the latter does not happen, but due to forecast uncertainties, explicit statements such as ”flood levels will reach 0.5 feet in your house” will be wrong. If by chance, flood levels do exactly reach 0.5 feet, that will have been a lucky forecast, very likely involving some skill, but the flood level could have reached 0.43 or 0.72 feet as well. This research presents a flood forecasting method that improves on traditional methods by directly incorporating uncertainty information into flood forecasts that still appear like forecasts people are familiar and comfortable with and understandable by them.

M. Tech. Business Administration / Financial distress and bankruptcy is one of the most significant threats to the going concern of many businesses, irrespective of their size and nature of operations. Research in corporate financial distress and corporate failure prediction dates back to the mid-sixties, and the bulk of the studies have been conducted within the context of highly developed market economies. Very little research has been conducted within the context of emerging markets, and using small and medium-sized firms. This therefore encouraged the author of this research report to conduct a study, applying a model specifically developed for emerging economies to predict financial distress of small and medium-sized South African listed firms. The main purpose of this study is to examine whether a model designed for financial distress prediction and credit scoring in emerging markets is reliable, and can be accurately applied in South Africa.

Thesis (MA)--Stellenbosch University, 2002. / ENGLISH ABSTRACT: Since the beginning of the economic reforms in 1978, the People's Republic of China (PRC) has attracted continuous interest from foreign investors, both in the form of foreign direct investment (FDI) and international trade, making the PRC the second largest host of FDI in the world. Despite occasional declines in foreign investment, foreign investors remain very interested in the long-term prospects for doing business in the country. The PRC's phenominal economic growth, large consumer market, the accession to the World Trade Organisation (WTO), and the government's commitment to open markets, economic reforms, and restructuring of the economy are amongst the factors that continue to attract foreign investment and trade. Despite the huge market potential and strong desire by foreign investors to do business in The PRC, the track record of foreign companies and investments in the country have at best been mixed. While some foreign investors have reaped large profits, others have failed to meet their performance targets. Foreign investors have faced a number of problems that are not market or trade related, but associated with economic, political and social trends and developments, including corruption, nepotism, crime, poor infrastructure, a depleted banking system, inefficient legal system, unemployment and poverty. Therefore, it is not surprising that many foreign investors are asking themselves whether the benefits of doing business in the PRC are worth the risks. In an increasingly uncertain and instable international trade and investment environment political risk assessment and management have become essential components of any profitable foreign investment strategy. Consequently, numerous political risk-rating agencies and a large number of both qualitative and quantitative risk assessment methods have emerged over the years. There is, however, neither general consensus regarding the definition of political risk nor a comprehensively systematic method of conducting political risk assessment. The definitions of political risk include a wide variety of indicators, ranging from governmental actions to all non-market developments. The number of methods available for political risk analysis range from informal, unsystematic assessments by a few individuals to formal, systematic, and sophisticated risk analysis models. There are, however, some similarities. The main objective of political risk analyses is to describe, explain, and forecast political conditions and events that affect the interests of foreign investors operating abroad or planning to enter a foreign market. In addition, political risk analyses attempt to forecast losses, and recommend means of managing the risk, avoiding the losses, and seizing the opportunities. Scenario planning is one of the qualitative methods used to analyse political risk. Scenario planning, however, differs from most other approaches as it does not try to accurately predict what will happen in the future or to provide the right tool for foreseeing the future developments, but to offer a range of possible futures. The underlying assumption is that the future cannot be forecast or predicted with certainty, but that the very process of thinking about the future and exploring the implications of possible future scenarios may have a profound impact on foreign investment and trade. Scenario planning is a method that provides insightful information necessary to understand, anticipate and respond to change and uncertainty in the future PRC. The development of four 20-year scenarios in this study demonstrates that the prospects for foreign investment can be both positive and negative. When the economy continues to grow strongly, and the government is able to maintain a stable environment and successfully implement the necessary changes foreign investors are expected to reap the desired benefits. However, if the problems facing the PRC at the moment further deteriorate foreign investors could expect increased risks, and the possibility of failure. / AFRIKAANSE OPSOMMING: Vanaf die begin van die ekonomiese transformasie in 1978, het die Volksrepubliek van Sjina voortdurende belangstelling van buitelandse beleggers geniet. Hierdie belangstelling was gemanifesteer in die vorm van direkte buitelandse belegging asook internasionale handel. Sjina het so aanloklik vir buitelandse beleggers geword, dat dit tans die wêreld se tweede grootste ontvanger van buitelandse belegging is, en beleggers stel veral belang in die lang termyn moontlikhede van besigheid doen in die land. Die Volksrepubliek van Sjina se merkwaardige ekonomiese groei, groot verbruikersmark, toetreding tot die Wêreld Handels Organisasie, en die regering se verbintenis aan die ontwikkeling van 'n oop ekonomie, ekonomiese transformasie en die herstrukturering van die ekonomie as sulks, is sommige van die faktore wat toenemend buitelandse belegging en handel aanlok. Ten spyte van die groot verbruikersmark potensiaal en die sterk begeerte van buitelandse beleggers om besigheid te doen in die Volksrepubliek van Sjina, is die ervarings van buitelandse maatskappye tot dusver gemeng. Alhoewel sommige buitelandse beleggers groot wins gemaak het, het ander minder sukses ervaar. Buitelandse beleggers word ook gekonfronteer met baie probleme wat nie noodwendig met die mark of handel gepaard gaan nie. Hierdie probleme word geassosieer met ekonomiese, politieke en sosiale gebeure en faktore insluitend korrupsie, misdaad, nepotisme, swak infrastruktuur, 'n ledige bank sisteem, 'n ondoeltreffende regssisteem, werkloosheid en armoede. Baie buitelandse beleggers betwyfel dus moontlik die vraag of besigheid doen in die Volksrepubliek van Sjina tog meer voordele inhou as risiko. In 'n wêreld waar internasionale handel en belegging met onsekerheid en onstabiliteit gepaard gaan, het die aspekte van politieke risiko skadebepaling en -bestuur belangrike komponente van enige winsgewende buitelandse belegging strategie geword. Gevolglik het verskeie politieke risiko-analise agentskappe asook 'n verskeie risiko-analise metodes van beide kwantitatiewe en kwalitatiewe aard hul verskyning gemaak. Ten spyte van die bogenoemde is daar nog steeds geen veralgemeende konsensus oor die definisie van politieke risiko, of 'n oorsigtelik sistematiese metode van politieke risiko-skatting onderneem nie. Die definisies van politieke risiko sluit in 'n groot verskeidenheid van indikatore wat wissel van regeringaksies tot mark-onverwante gebeure. Die verskeidendheid van metodologië wat gebruik word in politieke risiko-analise wissel van informeel, onsistematiese skattings, tot formele, sistematiese en gesofistikeerde risiko-analise modelle. Die primêre doel van politieke risiko-analise is om te beskryf en te verduidelik, en ook om politieke omstandighede en gebeurtenisse wat die belangstelling van buitelandse beleggers affekteer te voorspel. Addisioneel beoog politieke risiko-analise om ook die moontlikheid van verlies te voorspel en om 'n strategie vir die bestuur van die risiko aan te beveel, om sodoende verlies so ver moontlik te vermy asook om moontlike geleenthede aan te gryp. Senariobeplanning is een van die kwalitiatiewe metodes wat gebruik kan word in die analise van politieke risiko. Senariobeplanning verskil van ander benaderings in die sin dat dit nie akkurate voorspellings vir die toekoms as sulks maak nie, maar eerder 'n verskeidenheid van moontlike toekomstige omstandighede weergee. Die ontwikkeling van vier 20-jaar senarios vir die Volksrepubliek van Sjina in hierdie studie illustreer hoe die uitsigte vir buitelandse belegging positief sowel as negatief kan wees. Indien die Sjinese ekonomie verder groei en die regering in staat is om 'n stabiele omgewing in stand te hou asook die nodige veranderings te implimenteer, kan buitelandse beleggers verwag om beoogde voordele van buitelandse belegging te ervaar. Maar as die probleme wat die Volksrepubliek van Sjina op die oomblik ervaar voortduur en lof verswak, kan buitelandse beleggers verhoogde risiko sowel as die moontlikheid van mislukkings verwag.

Thesis (MScEng (Civil Engineering))--University of Stellenbosch, 2006. / The climate of the world varies from one decade to another, and a changing climate is natural and expected. However there is a well-founded concern that the unprecedented human industrial development activities of the past two centuries (and mainly the last century) have caused changes over and above natural variation. Climate change is the natural cycle through which the earth and its atmosphere are going to accommodate the change in the amount of energy received from the sun. There are various indicators that can be monitored to measure and verify possible climatic changes. This thesis will firstly emphasize what the possible effects of climate change could be on amongst others, the coastal zone, biodiversity and water resources. If the impact of climate change on the above mentioned processes are monitored, and changing trends can be identified, these processes could in fact be seen as climate change indicators. This is of major importance to us, to be able to accurately identify whether climatic changes are experienced in any given area and to attempt to quantify it. Engineering hydrologists are, amongst other duties, responsible for the determination of peak discharges to be able to size conduits to safely convey the stormwater for given recurrence interval events. All hydrological predictions are indirectly or directly based on historical data. Empirical formulas and deterministic methods were developed and calibrated from known historical data. Statistical predictions are directly based on actual data. The question that arises is whether the historical data still provides an accurate basis from which possible future events can be predicted? This thesis strives to find an answer to this question and will also try to advise hydrologists on how they should interpret historical data in the future, taking climate change into consideration. The methodology that will be followed will be to compare the percentage of occurrence of 24-hour rainfall events of different magnitudes, for historical- as well as predicted rainfall, for five different rainfall stations in the Western Cape. A detailed analysis of measured data at a rainfall station, with 42 years of useable data, will also be performed, to verify whether any measurable trends have already been experienced. Conclusions shall be drawn as to possible trends, and recommendations will be made as to how hydrologists could allow for the possible changing rainfall patterns.

published_or_final_version / Urban Planning / Master / Master of Science in Urban Planning

published_or_final_version / Business Administration / Master / Master of Business Administration

published_or_final_version / Statistics / Master / Master of Social Sciences

Effective supply chain management gains much attention from industry and academia because it helps firms across a supply chain to reduce cost and improve customer service level efficiently. Focusing on one of the key challenges of the supply chains, namely, demand uncertainty, this dissertation extends the work of Zhao, Xie, and Leung so as to examine the effects of forecasting method selection coupled with information sharing on supply chain performance in a dynamic business environment. The results of this study showed that under various scenarios, advanced forecasting methods such as neural network and GARCH models play a more significant role when capacity tightness increases and is more important to the retailers than to the supplier under certain circumstances in terms of supply chain costs. Thus, advanced forecasting models should be promoted in supply chain management. However, this study also demonstrated that forecasting methods not capable of modeling features of certain demand patterns significantly impact a supply chain's performance. That is, a forecasting method misspecified for characteristics of the demand pattern usually results in higher supply chain costs. Thus, in practice, supply chain managers should be cognizant of the cost impact of selecting commonly used traditional forecasting methods, such as moving average and exponential smoothing, in conjunction with various operational and environmental factors, to keep supply chain cost under control. This study demonstrated that when capacity tightness is high for the supplier, information sharing plays a more important role in effective supply chain management. In addition, this study also showed that retailers benefit directly from information sharing when advanced forecasting methods are employed under certain conditions.