Two principal results for reputation risk are established. First, reputation risk can be measured in terms of a single index, arising from a data mining process directed at the opinions in a complex multiagent network. Second, the results of the measurement process, gathered over an extended period, can be expressed directly in monetary terms by finding a correlation between the daily changes in the index and in sales. Stressed periods are modelled by calculating value-at-risk using a 'loss-distribution/ scenario' approach, as for operational risk capital. The short-term effect of reputation risk events on sales and profits can be significant in absolute terms, but is small as a percentage of total sales. Negative reputation has a more significant impact than positive reputation.
A black-box optimization problem is considered, in which the function to be optimized can only be expressed in terms of a complicated stochastic algorithm that takes a long time to evaluate. The value returned is required to be sufficiently near to a target value, and uses data that has a significant noise component. Bayesian Optimization with an underlying Gaussian Process is used as an optimization solution, and its effectiveness is measured in terms of the number of function evaluations required to attain the target. To improve results, a simple modification of the Gaussian Process ‘Lower Confidence Bound’ (LCB) acquisition function is proposed. The expression used for the confidence bound is squared in order to better comply with the target requirement. With this modification, much improved results compared to random selection methods and to other commonly used acquisition functions are obtained.
New goodness-of-fit measures which are significant improvements on existing mea-Changes to this sentence and the next OK? sures are described. They use the intuitive geometrical concept of the area enclosed by the curve of a fitted distribution and the profile of the empirical cumulative distribution function. A transformation of this profile simplifies the geometry and provides three new goodness-of-fit tests. The integrity of this transformation is justified by topological arguments. The new tests provide a quantitative justification for qualitative judgements on goodness-of-fit, are independent of population size and provide a workable way to objectively choose a best fit distribution from a group of candidate distributions.
Poor performance of artificial neural nets when applied to credit-related classification problems is investigated and contrasted with logistic regression classification. We propose that artificial neural nets are less successful because of the inherent structure of credit data rather than any particular aspect of the neural net structure. Three metrics are developed to rationalise the result with such data. The metrics exploit the distributional properties of the data to rationalise neural net results. They are used in conjunction with a variant of an established concentration measure that differentiates between class characteristics. The results are contrasted with those obtained using random data, and are compared with results obtained using logistic regression. We find, in general agreement with previous studies, that logistic regressions out-perform neural nets in the majority of cases. An approximate decision criterion is developed in order to explain adverse results.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.