This three-day course aims to provide participants with the tools necessary to conduct robust risk analyses using @RISK5 and the other products in the DecisionTools Suite, across a range of applications. The course aims to maximise hands-on experience by building most models from scratch and using this process to facilitate the learning and use of the software in practice.

Description

Day 1 is focussed on the fundamentals of @RISK, including learning to operate the software and to interpret the results of the analysis, including hands-on exercises related to cost budgeting, risk registers, and scenario modelling.

Day 2 is focussed on extending participants’ @RISK modelling techniques, including further topics in the selection and use of distributions and the capturing of dependency relationships, such as correlation. Further features of the software are also explored, and participants’ own modelling situations are discussed. Hands-on exercises include cash flow modelling, time series modelling, and the implementation of correlation.

Day 3 covers the other products in the DecisionTools Suite.  We cover PrecisionTree to perform decision-tree analysis, and RISKOptimizer and Evolver for optimisation problems.  We then discuss TopRank to support the auditing of models and to conduct sensitivity analysis, and use StatTools to perform a selection of statistical procedures, and NeuralTools to make predictions based on historic data.

Days 1 and 2 are identical to the stand-alone course “Decision-Making and Quantitative Risk Analysis using @RISK”.  Participants with previous knowledge of @RISK may attend day 3 only, at their discretion.

Instructor

The course is taught by Dr. Michael Rees, Palisade’s Global Director of Training and Consulting.  His prior experience includes 15 years as a strategy consultant, equity analyst and independent consultant.  He has a Doctorate in Mathematics and a B.A. with First Class Honours from Oxford University, and a MBA with Distinction from INSEAD.  He has studied quantitative finance with Paul Wilmott, graduating top of the class and also receiving the Wilmott Award in 2003.  He prides himself on his ability to explain complex concepts in simple, focused, intuitive, and non-mathematical ways.  He is the author of Financial Modelling in Practice (John Wiley & Sons, 2008).

Detailed Agenda

Day 1: Fundamentals of @RISK

AM

• Introduction to risk analysis
• Benefits and purpose
• Brief revision of key statistical terms
• Hands-on exercise: cost budgeting
• Introduction to risk modeling using distributions
• Exercise using Triangular distributions
• Viewing an @RISK model
• RiskStatic and @RISK function swap
• Running a simulation, viewing and interpreting results
• Repeating a simulation, random number methods and number of iterations required
• Hands-on exercise: Risk mitigation
• Use of multiple simulations
• Use of @RISK Statistics functions for model outputs and inputs

PM

• Hands-on exercise: scenario modeling
• Use of the Discrete distribution
• Comparison with multiple simulations
• Introduction to modelling event risks
• Hands-on exercise: risk registers and frequency-severity models
• Use of the Binomial distribution
• Use of the Poisson, and Compound distributions (overview)
• Related applications (oil and gas, sales forecasting)
• Dependency modeling example (incl. multiple simulations)
• Further features
• Excel reports
• Simulation data
• Working with @RISK graphs
  

Day 2: Building Robust Models with @RISK

AM

• Selection and use of distributions for continuous uncertainty
• The Triangular and PERT distributions
• The Normal and Lognormal distributions
• Alternate parameter methods
• Hands-on exercises: Cash flow, time-series and further modelling applications
• Cash flow and time-series modelling (incl. extensions, such as mean-reverting processes, crash processes, Markov chains, as relevant)
• Valuing flexibility and real options
• Extensions of cost and risk register models
• Working with data
• Distribution fitting with BestFit
• Re-sampling methods
• Hands-on exercise: Modelling time-to-occurrence
• Introduction to other distributions (Geometric, Exponential, Weibull etc.)

PM

• Dependency and correlation modelling
• Comparing parameter dependency with correlated sampling (meaning, advantages and disadvantages)
• Measuring correlation coefficients
• Consistency of correlation matrices
• Hands-on exercise: Implementing correlation
• In cost modeling
• In time series models
• Further aspects of results reporting
• Tornado graphs and scatter plots: Creation and interpretation (incl. common mistakes)
• Model auditing tools
• Further topics and examples (open session) e.g.
• Review of pre-built models (e.g. oil and gas, environmental, pricing, insurance, optimisation etc)
• Use of other sensitivity analysis tools, macros, other functions and features etc
  

Day 3: Introduction to other products in the DecisionTools Suite

AM

Session 1: Decision-making using PrecisionTree

• Introduction to PrecisionTree
• Hands-on examples (cumulative pay-off trees)
• Basic everyday example
• Drug-development example
• Testing example (as relevant)
• Sensitivity analysis with decision trees
• Introduction to other aspects of PrecisionTree
• Other example models
• Non-cumulative payoff trees
• Use with @RISK (overview)
• Other features

Session 2: Sensitivity Analysis and Optimisation using TopRank, Evolver and RISKOptimizer

• Introduction to TopRank and hands-on example (cash flow model)
• Finding model inputs
• Performing sensitivity analysis
• Introduction to Evolver to perform optimisation and hands-on example (factory capacity optimisation)
• Introduction to RISKOptimizer to perform optimisation and hands-on example (variation of factory capacity optimisation)
• Speeding up optimisation: tips

PM

Session 1: Model Auditing, and Predictive and Statistical Analysis using StatTools and NeuralTools

• Introduction to selected StatTools statistical procedures and hands-on examples:
• Correlation calculations
• Confidence intervals and hypothesis tests
• Normality and other tests
• Introduction to predictive modelling using NeuralTools (hands-on examples with numerical and categorical variables)

Session 2: Open Session (as per participants needs)

• Example topics
• Further review of @RISK models (pre-built models or participants’ models)
• Using DTS products in an integrated way
• Etc