American Option Pricing with Importance Sampling and Shifted Regressions

This paper proposes a new method for pricing American options that uses importance sampling to reduce estimator bias and variance in simulation-and-regression based methods. Our suggested method uses regressions under the importance measure directly, instead of under the nominal measure as is the standard, to determine the optimal early exercise strategy. Our numerical results show that this method successfully reduces the bias plaguing the standard importance sampling method across a wide range of moneyness and maturities, with negligible change to estimator variance. When a low number of paths is used, our method always improves on the standard method and reduces average root mean squared error of estimated option prices by 22.5%.

Ishikawa diagrams and Bayesian belief networks for continuous improvement applications

Purpose In continuous improvement (CI) projects, cause-and-effect diagrams are used to qualitatively express the relationship between a given problem and its root causes. However, when data collection activities are limited, and advanced statistical analyses are not possible, practitioners need to understand causal relationships. The paper aims to discuss these issues. Design/methodology/approach In this research, the authors present a framework that combines cause-and-effect diagrams with Bayesian belief networks (BBNs) to estimate causal relationships in instances where formal data collection/analysis activities are too costly or impractical. Specifically, the authors use cause-and-effect diagrams to create causal networks, and leverage elicitation methods to estimate the likelihood of risk scenarios by means of computer-based simulation. Findings This framework enables CI practitioners to leverage qualitative data and expertise to conduct in-depth statistical analysis in the event that data collection activities cannot be fully executed. Furthermore, this allows CI practitioners to identify critical root causes of a given problem under investigation before generating solutions. Originality/value This is the first framework that translates qualitative insights from a cause-and-effect diagram into a closed-form relationship between inputs and outputs by means of BBN models, simulation and regression.

Overheating risk and cooling demand in residential buildings: performance prediction and improvement using a prescriptive approach

Depending on general climatic trends and specific (urban) microclimatic conditions, the reliable estimation of the overheating risk in buildings has become increasingly important. As such, detailed simulation of the related phenomena can provide useful information. However, deployment of detailed simulation involves a number of challenges, including time and effort expenditures not accounted for in typical building delivery processes. In this context, careful application of prescriptive methods may provide – at least for a specific class of applications – a reasonable alternative. The present contribution explores this possibility via a specific case study involving a large sample of residential buildings in Gaza, Palestine. This sample includes some fifty multi-unit apartment buildings representing the bulk of residential building stock in Gaza. These buildings were assessed via both numeric simulation and regression-based methods, assuming the latter can provide the basis for development and validation of a prescriptive approach. Toward this end, a number of independent variables were considered and the level of their association with the computed values of the designated building performance indicators was observed. Comparison of the regression-based and simulation-based methods revealed a reasonable level of agreement. This suggests that the proposed prescriptive method may provide an attractive alternative to highly detailed simulation.