No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

Most lending institutions are aware of the importance of having a well-performing credit scoring model or scorecard and know that, in order to remain competitive in the credit industry, it is necessary to continuously improve their scorecards. This is because better scorecards result in substantial monetary savings that can be stated in terms of millions of dollars. Thus, there has been increasing interest in the application of new classifiers in credit scoring from both practitioners and researchers in the last few decades. Most of the recent work in this field has focused on the use of new and innovative techniques to classify applicants as either 'credit-worthy' or 'non-credit-worthy', with the aim of improving scorecard performance. In this thesis, we investigate the suitability of intelligent systems techniques for credit scoring. In particular, intelligent systems that use immunological metaphors are examined and used to build a learning and evolutionary classification algorithm. Our model, named Simple Artificial Immune System (SAIS), is based on the concepts of the natural immune system. The model uses applicants' credit details to classify them as either 'credit-worthy' or 'non-credit-worthy'. As part of the model development, we also investigate several techniques for selecting variables from the applicants' credit details. Variable selection is important as choosing the best set of variables can have a significant effect on the performance of scorecards. Interestingly, our results demonstrate that the traditional stepwise regression variable selection technique seems to perform better than many of the more recent techniques. A further contribution offered by this thesis is a detailed description of the scorecard development process. A detailed explanation of this process is not readily available in the literature and our description of the process is based on our own experiences and discussions with industry credit risk practitioners. We evaluate our model using both publicly available datasets as well as a very large set of real-world consumer credit scoring data obtained from a leading Australian bank. The evaluation results reveal that SAIS is a competitive classifier and is appropriate for developing scorecards which require a class decision as an outcome. Another conclusion reached is one confirmed by the existing literature, that even though more sophisticated scorecard development techniques, including SAIS, perform well compared to the traditional statistical methods, their performances are not statistically significantly different from the statistical methods. As with other intelligent systems techniques, SAIS is not explicitly designed to develop practical scorecards which require the generation of a score that represents the degree of confidence that an applicant will belong to a particular group. However, it is comparable to other intelligent systems techniques which are outperformed by statistical techniques for generating p ractical scorecards. Our final remark on this research is that even though SAIS does not seem to be quite suitable for developing practical scorecards, we still believe that there is room for improvement and that the natural immune system of the body has a number of avenues yet to be explored which could assist with the development of practical scorecards.

No description available.

The Moehau Kiwi Sanctuary is one of five sanctuaries established in 2000 and managed by the Department of Conservation. The objective of the sanctuaries is to protect the most endangered kiwi taxa, and increase kiwi survivorship. Operation Nest Egg (ONE) is a programme utilised by the Moehau Kiwi Sanctuary for artificially incubating abandoned Kiwi eggs and captive rearing chicks until they begin to show a gain in weight. ONE chicks were then released back onto Moehau or adjacent protected areas. Kiwi populations are declining on the mainland at an average of about 3% per year in areas where predators of kiwi are not controlled. The main cause for this decline is chick mortality due to predation by stoats (Mustela erminea). During natal dispersal kiwi are known to disperse significant distances of between 5–20 km, and this has influenced the size of management areas needed for the protection of kiwi (10,000 hectares). The type of forest-cover is an important element in determining where management areas are located, as kiwi has preferences for certain forest types over others. This study conducted at Moehau, Coromandel, on the North Island Brown Kiwi advances our knowledge of kiwi by examining differences in rates and distances of dispersal among chicks, sub-adults, non-territorial and territorial adults, as well as between genders. This study investigates kiwi selective use of roost site types, ground-cover types, forest types and physiographical features. Addressed in this study are differences in dispersal, habitat selection and mortality among age-classes and between genders over the months of the year, and across elevations. Comparisons between ONE and wild-reared kiwi dispersal and mortality are included. Data were collected between 2001 and 2008 from observations of kiwi located during daytime hours. The data recorded included the grid reference, elevation, ground-cover type, forest type, physiography, and the type of roost site. The sample size for this study was significantly larger than for any previous studies thus enabling a greater confidence in estimated dispersal rates and dispersal distances, habitat selection and factors relating to mortality. All wild-reared kiwi displayed dispersal and were not philopatric to their natal area. Dispersal distances were found to be further than previously estimated, with the net distance of natal dispersal differing among age-classes, from an average of 834m (SE +/- 131) for kiwi chicks to 7,553m (SE = +/- 1167) for non-territorial adults. Female sub-adult kiwi dispersed further (7,215m) than male sub-adult kiwi (4,226m) (p = 0.04). The time taken to travel one km during natal dispersal ranged from an average of 131days/km (SE = +/- 9) for chicks to 89 days/km (SE = +/- 13) for sub-adults. Habitat selection has been observed in other studies on kiwi but not specifically for Coromandel North Island Brown Kiwi, and selection for ground-cover types by kiwi when roosting on the surface has never been previously studied. Roost site selection of kiwi differed among age-classes (p <0.001), between gender (p <0.001), and across elevations (p <0.001). Female kiwi were found more often in surface roosts (64%) than hole roosts (32%), and male kiwi were found at similar frequencies in holes (46%) and on the surface (47%). Sub-adults used holes to a greater extent as elevation increased, and selected for sub-alpine forest over broadleaf forest (p <0.001). This study is the first to recognise that selection of ground-cover types by kiwi differs among age-classes (p <0.001). Kiwi chicks were more often found on the surface under dead fern fronds and debris (39%) than other ground-cover types. The mortality rate was highest in chicks (33%), with predation responsible for 60% of these deaths; conservation management techniques were responsible for a further 20% of deaths; the remaining 20% of deaths were due to natural or unknown causes. Summer (December-February) was the season in which 81% of kiwi chick deaths occurred. The high proportion of deaths from monitoring techniques and the use of radio-transmitters (22%) indicates improvements need to be made to current management practices. ONE chicks were found to disperse shorter distances and had a greater mortality rate than wild-reared chicks. Therefore, recommendations are made for changes to ONE management practices. Further recommendations are made for the enhancement of kiwi habitat that could reduce kiwi mortality, and for increasing the habitat available to kiwi, thereby potentially increasing population sizes and/or densities.

All venture capital investments exhibit some form of asymmetric information. The seminal paper on the structure of venture investments, Kaplan and Stromberg (2004), investigates how venture capitalists use deal construction to control agency conflicts within funded deals and their associated internal, external and execution risks. Another key strand of the academic literature has reviewed the contractual arrangements venture capital firms reach, the process of venture capital selection and determinants of their success from a post-investment perspective (Fried and Hisrich (1994), Manigart, Vermeir and Sapienza (1996), Gompers and Lerner (2004), Wright and Robbie (1998)). This thesis also explores venture capital investment, albeit from a preinvestment standpoint. In contrast to Kaplan and Stromberg???s (2004) demonstration of the use of venture capital mechanisms to control agency issues, this research addresses how agency issues influence the final selection of potential investments by venture capitalists. Kaplan and Stromberg (2004) use post-funding metrics to capture risks, which influence post-contract design. From a pre-funding perspective, internal, external and execution risks are subjective, rare and difficult to measure. Nevertheless, this thesis uses pre-funding proxies to replicate these risks, some of which have direct empirical academic support. Information for sixtytwo deals, thirty-four funded and twenty-eight unfunded, was hand collected through a combination of surveys, interviews and consultation with five of Australia???s leading venture capital firms, and individuals from the Australian Private Equity and Venture Capital Association (AVCAL) board and executive. The key results indicate that once past initial screening stages, investment proposals that have a higher likelihood of receiving venture investment are those that had prior government investment, and/or, where the entrepreneur has proposed the investment be through milestone tranches and where revenue is already being generated (for early stage ventures). The results suggest that venture capitalists tend to allocate capital to investments perceived as ???safer??? with respect to agency conflicts. More specifically, venture capitalists are more reliant on signals of quality and lower risk, such as government grants, restriction of capital outlay and prior revenue generation ??? all of which reduce associated levels of internal and execution risk in new ventures.