KEYNOTE: Modelling Uncertainty: The Good,The Bad & The Ugly

Dr. Dale F Cooper
Broadleaf Capital International

Modelling uncertainty is difficult. Tools like @RISK are easy to use, but getting reliable and justifiable outcomes takes more than a good tool. This talk will address some of the modelling pitfalls we see in our risk consulting activities, explore how and why they arise, and suggest some solutions. The focus will be on the way models are intended to support decision making and their effectiveness in doing this.

» Download Now

Building a Better Process to Support Executive Decision Making

Jay Horton
Strategis Partners

Many senior executives know only two degrees of probability, zero and one, while business analysts know every degree of probability except zero and one. So how do we bridge the knowledge gap? Quantitative decision support software and techniques have gone through a remarkable development over the past 20 years. However this has not necessarily led to sustained improvements in the quality of decision making in companies that have adopted these tools. Jay’s talk will draw upon his 20 years of experience consulting to senior executives on business decisions to explain how you can build a better process to support executive decision making. This involves understanding of the realities of how organisations make decisions, anticipating the needs of the decision maker, and overcoming inherent biases in decision making.

» Download Now

Modeling Correlation and VAR of Swaps

Dr. Nitin Singh
Indian Institute of Management Indore

Industry Focus: Finance
Product Focus: @RISK, StatTools

This presentation will show how swap was valued and how the volatility of swap was measured through simulation using @RISK. We have also proceeded to perform initial diagnostics on time series data and computed variance-covariance matrix using StatTools. As a spin-off, we learned how volatility of swap depends on the variance-covariance structure of the zero prices. We begin the analysis by estimating zero-coupon bond prices: First we consider cash flows of the swap in different time periods. Next, we value the swap by computing the present value of each leg and taking the difference between two values. The value of the floating leg is determined since the next floating coupon is known and we dealt with generic interest-rate swap. The present value of the floating-leg is calculated by pricing the swap over time. Thereafter, VAR is calculated using simulation by @RISK. We find cumulative probability for each change in Call Value and compute VAR for a 99% confidence level. Interestingly, we find that volatility of swap and VAR depends on the variance-covariance matrix of the zero prices.

» Download Now

Modelling Sheep Population and Wool Production in Australia

Kimbal Curtis
Department of Agriculture and Food, Western Australia

Industry Focus: Agriculture
Product Focus: NeuralTools, @RISK

Many business planning decisions are based on forecasts of future industry conditions. The sheep and wool industry in Australia and the down stream wool processors around the world are no different. A key input to their decision making is a forecast of expected wool production, and as Australia dominates the supply of traded raw wool, Australian forecasts are of utmost importance.

Single point forecasts do not assist businesses to assess risk forcing contingencies to be adopted. Provision of confidence limits allows assessment of both lower and higher than expected production.

This presentation will outline a simplified deterministic flock demographic model of the Australian sheep population. The process undertaken to convert it to a stochastic model capable of improving forecasts will be described. Sensitivity analyses were used to identify which inputs to vary, and where most effort should be applied in collecting reliable input statistics.

» Download Now

Real Options Valuation in Property Leases and Deriving Value from Inherent Volatility

Dr. Colin Beardsley
Risk-Return Solutions Pty Limited

Industry Focus: Real Estate
Product Focus: @RISK

Real options analysis draws from financial markets by developing a framework for valuing flexibility in decision making which adds to traditional discounted cash flow (DCF) methodology. This case study uses @RISK Monte Carlo simulation to value the real options inherent in long term property leases and how volatility can increase real estate value.

» Download Now

Real-World Applications of NeuralTools

NeuralTools is a powerful Excel add-in that uses neural network technology to make powerful predictions in any spreadsheet model.  The presenters will cover three specific real-world case studies where NeuralTools has been used effectively: in business, in medicine, and in disaster response. 

Kimbal Curtis will discuss how NeuralTools predicted the market price of wool for the Western Australia Department of Agriculture.  Mark Meurisse will cover the use of NeuralTools for cardiovascular disease risk assessment.  He will also describe how NeuralTools was used to predict call volume for a disaster response center after Hurricane Katrina.

» Download Now: Kimbal Curtis examples

» Download Now: Mark Meurisse examples

Risk Analyses on a Wind Farm Project

William Zhu
WEL Networks Ltd

Industry Focus: Energy
Product Focus: @RISK, PrecisionTree

A wind power generation project is valuated based on Discounted Free Cash Flow (DFCF) method. Key risk factors, including revenue associated (e.g., the amount of electricity generated by wind and electricity spot prices), capital expenditure associated (e.g., the NZ exchange rate linked to the imported wind turbines), and operating expenditure associated (e.g., CPI and PPI fluctuation), are identified. By utilising @RISK software, the model discusses the following issues in detail:

• Distributions for the key input factors are fitted (especially for wind speed, electricity spot prices, and NZ exchange rates)
• Correlation constraints are added for some of the risk factors to capture the inter-relationships among them
• Both deterministic and stochastic sensitivity analyses are compared and discussed
• Different scenarios are investigated by using advanced @RISK techniques (including goal seek, stress, and sequential analyses)
• The scenario results are summarized by using a simple decision tree diagram.

This model uses Excel Macros, which incorporates numerous @RISK functions, to display the results and plot the graphs, so that users can run the model smoothly with little @RISK knowledge. It is also easy to convert this case study into a general business valuation model with minor modification.

» Download Now

Some Challenges and Solutions in the Application of Monte Carlo Risk Modelling to Health Risk Assessment in the Water Industry

Dr. David Roser
University of New South Wales, Centre for Water and Waste Technology

Industry Focus: Environment
Product Focus: @RISK

Quantitative risk techniques such as quantitative microbial risk assessment ( QMRA) are being increasingly applied to environmental and health impact/risk assessment. The introduction of QMRA is being driven by new inter/national guidelines, the availability of modelling tools like @RISK, and the need for assessments to account for the variability and uncertainty in environmental data such as water quality measurements. The CWWT uses QMRA to support its water and waste management projects and has found that QMRA risk modelling has a range of benefits. It facilitates an ‘holistic’ integration of hazard, dose response, and exposure pathway assessments central to health risk characterisation. Hazard ‘Exposure Pathway’ construction with QMRA in mind leads automatically to the identification of essential submodel inputs and major data gaps, and hence priorities for new experimental work and literature searches.

CWWT has constructed large numbers of models varying slightly but significantly in their assumptions, necessitating the development of ‘metamodelling’ and rigorous data management. Simulation of the impacts of infrequent ‘hazardous events’ has at times been impractical because of the large numbers of simulations required. And the visual complexity of CWWT models makes communication of modelling details to senior managers challenging. This presentation will outline how we have undertaken health risk modelling with @RISK using QMRA and water industry examples.  We will discuss its strengths and identify how to address the above constraints both in model construction and using @RISK.

» Download Now