
Schedule Announced
The program schedule for the 2008 Palisade Risk & Decision Analysis Conference is now available. The Conference is a two-day forum covering risk, decision, and data analysis from all angles. Full of real-life case studies using @RISK and the DecisionTools Suite as well as hands-on software training, the event will show you new approaches to the problems you face every day. You will also learn about to apply the all-new DecisionTools Suite 5.0 to a variety of situations.
» See the Conference Program
» Register now and Save US $200
» Watch testimonials from last year’s conference on YouTube
The Hyatt and Manhattan
The Palisade Risk & Decision Analysis Conference will take place at the gorgeous Hyatt Regency Jersey City on the Hudson. The hotel offers stunning views of New York City. Join us for cocktails and a dinner cruise around Manhattan and the Statue of Liberty, embarking from the Hyatt.
» Book your room now for only US $309
Palisade Partners with
Six Sigma Consultancy Rath & Strong
Palisade is pleased to announce the formation of a partnership with the management consultancy Rath & Strong, based in Lexington, Massachusetts. Rath & Strong is an acknowledged leader in the field of change management and counts many Fortune 500 companies among its clients. The firm’s current consulting and training emphasizes Lean Six Sigma, and the goal of its new partnership with Palisade is to expand the role of Monte Carlo simulation in Lean Six Sigma practice. Palisade’s software @RISK 5.0 will be an important analytic tool for Rath & Strong, and Palisade will advise the firm’s consultants on the use of Monte Carlo simulation in process improvement.
» Read the full press release
» Quality Digest recently covered the partnership
Free Live Webcasts
Mitigating Project Specification Risk using @RISK
Presented by Dr. William McKibbin
Global Director of DFSS, Bausch&Lomb Inc.
» 21 August 2008, 11:00 AM – Noon ET
What we will discuss in detail in this free live webcast are the kinds of projects contractors bid on (fully-specified versus quasi-specified), typical issues that arise, applicable distributions, and then, a case study application.
Data Analysis and Statistics for Decision Making Using StatTools 5.0
Presented by John McKenzie
Founder, Innovastat Corporation
» 28 August 2008, 11:00 AM – Noon ET
We are awash in data but sometimes do not use it to its full potential for the analysis of decisions. Innumeracy and statistical illiteracy hampers the analyst’s ability to think critically when data is part of the decision making process. John McKenzie will demonstrate how StatTools can be easily used for managing your data set, deriving descriptive statistics, exploratory data analysis, determining normality, inferential statistics (both parametric and non-parametric), time series, regression, logistic regression and quality control. He will use some interesting data sets from historical events and from scientific experiments in the explanation of the StatTools functions.
» Learn more about webcasts and register
VAIR Financial Modeling Courses
Featuring @RISK
Financial Statement Modeling and Analysis with Excel
and @RISK 5.0 - September 8-9 New York City
This class demonstrates efficient model creation and how these financial models should guide commercial discussions. Participants learn how to "negotiate off the financial model." This course is indispensable to the effective use of financial spreadsheets for any industry or sector.
» More information or Register

Can a Single Model Replace Simulation in Determining Optimal Portfolio Withdrawal Rates from a Retirement Portfolio?
Presented at the 2007 Risk & Decision Analysis Conference
Dr. Michael Tucker
Professor of Finance
Fairfield University, Dolan School of Business
Milevsky and Robinson (2005) have derived a formula using the gamma distribution to estimate the probability of running out of money during retirement. The initial assumptions are a fixed withdrawal rate in real dollars for the retirement portfolio. What is uncertain is the performance of the underlying assets. @RISK is used to gauge the probability of running out of money under uncertainty. An Excel macro linked to @RISK calculates the probability of ruin at each withdrawal rate for each age and life expectancy.
» Download the presentation
» Download the model
TopRank and @RISK:
Product Launch
TopRank recognizes @RISK distribution functions and incorporates them in What-If analyses. This ability provides more flexibility and accuracy in modeling the possible input values in your What-If analysis. In this example, Jupiter Corporation is building a new model of 4-door sedan. Assuming that the car will generate sales for the next 5 years, management has identified 5 factors that can influence the total revenue during that period. Several of these factors have probability distributions associated with them. During a What-If analysis, TopRank sees the probability distributions associated with these items and performs a smart sensitivity analysis using them, stepping through the range of the distribution while spacing the steps such that each interval encompasses equal amounts of probability.
» Download example model: ProductLaunchTopRankRISK.xls
Oil and Gas Production Forecasting
This is a simple model forecasting production for a particular oil well. The estimated reserves within the well are uncertain and are represented with a Lognormal distribution function. The mean is 500,000 STB and the standard deviation is 50,000 STB.
The output in the model is the NPV of the reserves for the first 10 years of production. Other factors considered include the decline rate, the gas-oil ratio (GOR), the prices of oil and gas, as well as the rate of increase of the prices of oil and gas. The only input factor that contains an @RISK probability distribution function is reserves, but you could make the model more realistic by using distribution functions to describe the decline rate, GOR, price of oil, etc.
» Download example model: OilProdForecast.xls

Application Settings
@RISK 5.0, and the entire DecisionTools Suite 5.0, contains an Application Settings dialog where you can specify customizable defaults such as graph settings, window configurations, simulation settings, and more. This is located on the Utilities menu.
» See how Application Settings works

The Discrete Distribution
This is a user-defined distribution in which the user specifies all possible
outcomes and their probabilities. It can be used where it is believed there are
several discrete outcomes (e.g. worst case, expected case, best case), to
replicate some other discrete distributions (such as the Binomial distribution),
and to model discrete scenarios.
RiskDiscrete({X1,X2,...,Xn},{p1,p2,...,pn}) specifies a discrete distribution with
a number of outcomes equaling n. Any number of outcomes may be entered.
Each outcome has a value X and a weight p which specifies the outcome's
probability of occurrence. As with the RiskHistogrm function, weights may
sum to any value — they are normalized to probabilities by @RISK.
Examples
RiskDiscrete({0,.5},{1,1}) specifies a discrete distribution with 2 outcomes
valued 0 and .5. Each outcome has an equal probability of occurrence as the
weight for each is 1. The probability of 0 occurring is 50% (1/2) and the
probability of .5 occurring is 50% (1/2).
RiskDiscrete(A1:C1,A2:C2) specifies a discrete distribution with three
outcomes. The first row of the worksheet — A1 through C1 — holds the
values of each outcome while row 2 — A2 through C2 — holds the probability
“weight” of each occurring.
Guidelines Weight values p must be greater than, or equal to, zero and the sum of all
weights must be greater than zero.

Analytics Training
@RISK Demonstrates Financial Statement Modeling & Analysis
Technique: Monte Carlo simulation » see @RISK 5.0
Source: OilVoice, August 6, 2008
A new VAIR training course uses Palisade’s flagship software to teach the concepts and applications of Monte Carlo simulation.
Emergency Preparedness
Weighing In on Coastal Protection
Technique: Decision analysis » see DecisionTools Suite 5.0
Source: Daily Comet, Thibodaux, LA, August 1, 2008
Residents of Louisiana get hands-on experience using multivariate analysis to help the Army Corps of Engineers officials set priorities in case of future weather disasters.
Finance
A.M. Best Rates Collateralized Debt Obligations
Technique: Monte Carlo simulation » see @RISK 5.0
Source: MarketWatch, August 6, 2008
A worldwide insurance-rating and information agency used Monte Carlo simulation to evaluate the complicated packages of debts known as CDOs that are currently on the market.
Medicine
Predicting Treatment Outcomes for Hepatitis
Technique: Decision trees » see PrecisionTree
Source: 7thSpace Interactive, August 9, 2008
A new diagnostic model uses gene expression to predict the outcome of treatments for hepatitis C.
FDA Clears Monte Carlo Dose Calculation Algorithm
Technique: Monte Carlo simulation » see @RISK 5.0
Source: BioOptics World, July 21, 2008
The Monte Carlo algorithm is widely accepted as the gold standard for calculation of dosages. Now a specialized version of the algorithm allows highly precise radiation treatments.
Online Retail
Logistics: The Key to e-Retail Success
Technique: Neural networks » see @RISK 5.0
Source: Supply Chain Management Review, July 24, 2008
In order to facilitate the movement of goods purchased online, neural network pattern recognition, and other advanced data management techniques will have to be blended.
Personal Finance
Nest Egg Nostradamus
Technique: Decision analysis » see DecisionTools Suite 5.0
Source: Pioneer Press, St. Paul, MN, August 8, 2008
Caught with two mortgages in the housing slump, Dan and Nancy Boxrud turned to one of many financial advisers offering dedicated software for decision analysis.
Project Management
Assessing Overall Risk when Uncertainties are Linked,
by using Decision Trees
Technique: Decision trees » see PrecisionTree
Source: TechRepublic, July 29, 2008
Project managers can use probability-based decision trees to assess a series of related risks.
Psychology
Anorexia Can Take 25 Years Off Life
Technique: Decision analysis » see DecisionTools Suite 5.0
Source: National Post, Toronto, July 17, 2008
Canadian researchers used decision analysis software to analyze mortality statistics for 954 anorexia patients in BC over a 20-year period.
Robotics
Artificial Intelligence Has Potential to Overcome Human Handicaps
Technique: Neural networks » see NeuralTools 5.0
Source: Medical News Today , July 24, 2008
An interdisciplinary team of European researchers has developed a robotic arm and hand that moves and 'senses' like a human and is controlled by a neural network, a human-inspired electronic 'brain'.>
Robots Train Themselves to Move
Technique: Neural networks » see NeuralTools 5.0
Source: BBC News, August 6, 2008
Researchers in Leipzig have developed software that allows robots use trial and error to “learn” to move.

The Palisade User Forums are online discussion boards where users are invited to post questions and share ideas on their use of @RISK and other Palisade software. It’s also a great place to check for announcements regarding updates of Palisade software. Forums are organized by products.
» Join or view the Palisade Forums
Recent topics include:
» Serial correction of inputs |