Doctor of Philosophy / Department of Agricultural Economics / Allen M. Featherstone / Bryan W. Schurle / The reliability of nine software packages using the maximum likelihood estimator for the logistic regression model were examined using generated benchmark datasets and models. Software packages tested included: SAS (Procs Logistic, Catmod, Genmod, Surveylogistic, Glimmix, and Qlim), Limdep (Logit, Blogit), Stata (Logit, GLM, Binreg), Matlab, Shazam, R, Minitab, Eviews, and SPSS for all available algorithms, none of which have been previously tested. This study expands on the existing literature in this area by examination of Minitab 15 and SPSS 17. The findings indicate that Matlab, R, Eviews, Minitab, Limdep (BFGS), and SPSS provided consistently reliable results for both parameter and standard error estimates across the benchmark datasets. While some packages performed admirably, shortcomings did exist. SAS maximum log-likelihood estimators do not always converge to the optimal solution and stop prematurely depending on starting values, by issuing a ``flat" error message. This drawback can be dealt with by rerunning the maximum log-likelihood estimator, using a closer starting point, to see if the convergence criteria are actually satisfied. Although Stata-Binreg provides reliable parameter estimates, there is no way to obtain standard error estimates in Stata-Binreg as of yet. Limdep performs relatively well, but did not converge due to a weakness of the algorithm. The results show that solely trusting the default settings of statistical software packages may lead to non-optimal, biased or erroneous results, which may impact the quality of empirical results obtained by applied economists. Reliability tests indicate severe weaknesses in SAS Procs Glimmix and Genmod. Some software packages fail reliability tests under certain conditions. The finding indicates the need to use multiple software packages to solve econometric models.

Direct response advertising in the financial services industry in South Africa has become one of the most important tactics companies utilise to build and maintain market share. Ensuring that these advertising campaigns yield optimal return on investment numbers is the responsibility of marketing departments and their partners in the marketing and sales processes, such as the creative and media agencies, the distribution force, as well as the client service area that supports the client value proposition. The marketing executive therefore is accountable for the planning, budgeting and execution of direct response campaigns, which need to deliver sufficient results to support the company’s overall business objectives. The challenge all marketers face is the lack of a proven structured and scientific methodology to facilitate this planning, budgeting and execution process. It has always been a general view in the marketing fraternity that it is extremely difficult if not impossible to combine creative output measures, which are subjective in nature, with cost, sales and profit measures, which are objective in nature. This study aims to create a structured approach to marketing strategising and planning, by creating a marketing metrics model that enables the marketing practitioner to budget according to output needed to achieve the overarching business objectives of sales, cost management and profit. This marketing metrics model therefore unpacks the business drivers in detail, but through a marketing effort lense, to link the various factors underlying successful marketing output, to the bigger business objectives. This is done by incorporating both objective (verifiable data, such as cost per sale) and subjective variables (qualitative factors, such as creative quality) into a single model, which enables the marketing practitioner to identify areas of underperformance, which can then be managed, tweaked or discontinued in order to optimise marketing return on investment. Although many marketing metrics models and variables exist, there is a gap in the combination of objective and subjective factors in a single model, such as the proposed model, which will give the marketer a single tool to plan, analyse and manage the output in relation to pre-determined performance benchmarks. / Business Management / DCOM (Business Management)

Emerging insurance markets, have limited access to financial instruments that they can use to create common hedge(s) to manage foreign exchange risk. This is the first empirical study to focus on the limitations when modelling foreign exchange rate transaction exposure in emerging insurance markets. This work is based on the cash flow methodology proposed by Martin and Mauer (2003, 2005) in reference to banks, and employed by Li et al. (2009) when assessing US insurance companies. Some econometric methodological innovations have been introduced to study the limitations of modelling foreign exchange rate transaction exposure in emerging insurance markets. An extensive literature review is followed by a quantitative investigation, to answer the following research questions. 1) Is the foreign exchange transaction exposure, as measured by a fundamental (economic) method of modelling the interplay of foreign exchange rates with other economic variables, significant, for all Egyptian insurance companies? 2) Is the foreign exchange transaction exposure, as measured by a technical (statistical) way of modelling the interplay of foreign exchange rates with other economic variables, significant for all Egyptian insurance companies? 3) Is the exchange transaction exposure for the Egyptian insurance industry, as a whole, significant? Although the foreign exchange rate transaction exposure for the Egyptian insurance industry, as a whole, is insignificant (question3), the percentage of Egyptian insurers affected by foreign exchange rate transaction exposure in US dollars, estimated at the individual firm level, was found to be 22% (question 1) and 35% (question2) respectively.

Thesis (MBA (Business Management))--University of Stellenbosch, 2009. / ENGLISH ABSTRACT: Tobin's q value is widely used by financial analysts as a performance indicator ratio. The market value of a firm over the replacement cost of fixed assets and inventory serves as an indication of whether value is created by investing internally in the firm, or whether value is destroyed by investing in negative net present value projects. Where Tobin's q is greater than one (q > 1), the market value of the firm is greater than what it would cost to replace fixed assets and inventory. Therefore value is created. Firms that have a Tobin's q value of less than one are advised to pay dividends rather than invest in negative net present value projects. Over 200 different methods exist of calculating Tobin's q. By increasing the complexity of the algorithm to determine q, very little is achieved to improve the measurement quality. A strong link exists between excess market returns, free cash flow spending announcements and Tobin's q value for the firm. Firms with a high Tobin's q value should ensure that good investment possibilities are pursued. The use of internal funds to fund new investment is viewed in a positive light by the market and above average returns are generated. Firms with a high Tobin's q value and high free cash flow show lower returns. These lower returns happen as a result of the market recognising the firm's failure to capitalise on favourable internal investment opportunities. / AFRIKAANSE OPSOMMING: Tobin se q-waarde word wyd gebruik as prestasie aanwyser deur finansiele ontleders. Die markwaarde van 'n firma gedeel deur die vervangingskoste van vaste bates en voorraad, dien as 'n maatstaf om aan te dui of waarde geskep word deur intern in die firma te belê en of waarde vernietig word deur in projekte met 'n negatiewe netto teenswoordige waarde te belê. Waar Tobin se q-waarde groter is as een (q > 1) is die markwaarde van die firma groter as wat dit sal wees om die vaste bates en voorraad te vervang. Sodoende word waarde geskep. Firmas met 'n q-waarde van minder as een word aanbeveel om eeeder dividende uit te betaal as om die beskikbare fondse in projekte met 'n negatiewe netto teenswoordige waarde te investeer. Meer as 200 verskillende metodes bestaan om Tobin se q-waarde te bereken. Deur die kompleksiteit van die algoritme te vergroot om q te bereken, dra min by tot groter akkuraatheid van die meting. 'n Sterk verband bestaan tussen bo-gemiddelde markopbrengste, aankondigings oor die besteding van vrye kontantvloei en die Tobin q-waarde van die firma. Firmas met 'n hoë Tobin q-waarde moet verseker dat goeie investeringsgeleenthede aangegryp word. Die gebruik van interne fondse om nuwe investering te finansier word deur die mark in 'n positiewe lig beskou en bogemiddelde opbrengste word gelewer. Firmas met 'n hoë Tobin q-waarde en hoë vrye kontantvloei toon laer opbrengste. Hierdie laer opbrengste is as gevolg van die mark wat besef dat die firma nalaat om gunstige interne investeringsgeleenthede te gebruik.

Merit-based financial aid has long been utilized by college and university enrollment managers to attract the most academically qualified applicants for admission. Considerable research has been done to illustrate the impact of state-based merit aid programs and other scholarly pursuits have drawn attention to the consequences of merit aid on institutional investments in need-based aid. Less is known about the efficacy of merit aid to achieve college student enrollment objectives. The purpose of this study was to evaluate the relationship between merit aid values and the likelihood of undergraduate student enrollment yield on offers of admission. The primary research question to be answered was: What is the relationship between the amount of merit aid students receive from a college or university and their enrollment decisions? The sample comprised 2,770 students at three private higher education institutions in the United States. Binary logistic regression and a forward selection process were used to test a range of possible predictors (e.g., sex, race, ethnicity, in-state residency, distance from home, academic qualifications, merit aid awards, and information from the financial aid applications of those offered admission) to determine the relative strength of merit aid in the prediction of student enrollment yield on offers of admission. The amount of merit aid offered was positively related to the likelihood of a student to enroll, even when academic qualifications and other student characteristics were controlled.

published_or_final_version / Computer Science and Information Systems / Doctoral / Doctor of Philosophy

published_or_final_version / Real Estate and Construction / Doctoral / Doctor of Philosophy

published_or_final_version / Real Estate and Construction / Master / Master of Science in Real Estate and Construction

published_or_final_version / abstract / toc / Real Estate and Construction / Doctoral / Doctor of Philosophy

published_or_final_version / Economics and Finance / Master / Master of Philosophy