:: Live Web Training
Risk and Decision Assessment using @RISK, Part I
Risk and Decision Assessment using @RISK, Part II
Free Live Webcasts
Overview of @RISK for Project
Selecting the Right Distribution
Data Collection with
@RISK for Project
Can I determine how many iterations I need to run in my simulation so that the estimate of the mean is calculated within a specific confidence interval?
Yes you can. In the example here, let’s suppose that we want to use simulation to estimate the mean of the output in cell B11 and be accurate within 10 units 95% of the time. The number of iterations needed to meet these requirements can be calculated using the following formula:
n=[za/2S/D]2In this formula,
Palisade Announces Free Live Webcasts
Palisade is pleased to announce a new way to get more from your investment in @RISK and DecisionTools software: Free Live Webcasts. Webcasts are seminars given online free of charge to further learning in risk, decision, and data analysis. The sessions are 30 to 60 minutes in length, include question-and-answer, and are presented on a variety of topics. Webcasts will include a variety of different presentation formats, including:
:: Big Decisions, Big Models
Dr. Broskowski’s models are big simulations spun out over long time spans. Having first used @RISK in the early ‘90s for applying managed care principles to carving out mental health insurance benefits for top corporations like Federal Express and Chrysler, he has come to rely on its ease of use for even the largest models. In fact, although he now frequently trains his clients in the use of @RISK, he confesses that for many years he made little mention of his use of Palisade’s software. “I didn’t want those other guys—my competitors—to find out how easy it is to use.”
:: Balancing Risk
“With @RISK,” Dr. Broskowski says, “the real challenge isn’t the modeling. The real challenge is getting the folks in government to accept their role in risk sharing. Here again, @RISK is so intuitive it makes it easy to show people how to balance the risks with the opportunities and come to some kind of equilibrium.”
NeuralTools, Evolver and Solver
NeuralTools is Palisade’s Neural Networks add-in for Microsoft Excel that features the innovative Live Prediction function. Neural Networks are a rapidly-expanding approach to data analysis and prediction, but one that has been implemented primarily by analysts with an advanced understanding of data analysis and artificial intelligence. NeuralTools offers the power of sophisticated analysis, as well as instant updating with Palisade’s proprietary Live Prediction, to the everyday spreadsheet user.
:: Live Prediction a Standout
Live Prediction is the feature that makes NeuralTools a standout among all Neural Networks software. When new data are entered into an analysis, Live Prediction updates predictions instantly. This is a time-saver and is particularly valuable when combined with Excel’s Solver add-in or other optimization software such as Palisade’s Evolver.
:: Optimize, Predict, Repeat
Using Solver or Evolver, users can set a desired target prediction—for example, timely repayments of a bank loan with 90 percent certainty. Under a typical scenario, the optimizer would vary input variables such as loan amount until it achieved a 90 percent certainty of loan repayment. But with NeuralTools Live Prediction, the optimizer can vary loan amount, loan term, and other variables, and run that information through NeuralTools in real time. Thus as the optimization proceeds, NeuralTools continually updates its repayment predictions. Users can set the optimizer to continually change the model until the desired NeuralTools prediction is achieved. This is a far more accurate way of analyzing and optimizing predictive problems.
When coupled with Evolver, you can solve complicated, non-linear predictive problems. Combine Evolver’s genetic algorithms with NeuralTools’ predictive capabilities for unprecedented solving power.
:: Real-World Applications
The use of Neural Networks spans a variety of industry sectors as wide as the use of Excel—from banking to biology, defense to finance, and medicine to manufacturing. Designed for the many nonspecialists in all enterprises who want access to Neural Networks technology, NeuralTools has innumerable applications, including loan underwriting, credit card fraud detection, airline security analysis, medical diagnoses, financial investment prediction, and many more.
:: How NeuralTools Works
NeuralTools uses advanced statistical analysis techniques to reveal the structure of a data set. It “learns” existing data and then uses the knowledge and patterns it discovers to make predictions from new, unanalyzed data. All this happens “behind the scenes,” so the user need not be involved in setting the parameters of each algorithm. The software can perform both numeric and categorical prediction.
User Conference Big Success continued from above
The event drew professionals from all industries and featured two days of real-world case studies, workshops, networking, and a preview of the all-new @RISK 5.0 and DecisionTools Suite 5.0.
“The feedback we’ve received from the conference delegates has been overwhelmingly positive,” noted Sam McLafferty, President and CEO of Palisade Corporation. “Attendees expressed excitement about the new versions of @RISK and the DecisionTools Suite, and appreciated learning new applications from each other.”
:: @RISK 5.0 and Keynote Among Highlights
Considered by many a high point of the conference was Mr. McLafferty’s unveiling of @RISK 5.0 and the DecisionTools Suite 5.0. Spontaneous declarations of approval were heard as he explained feature after feature of the new versions, including: a more robust graphics engine, direct integration with the spreadsheet, @RISK Libraries for sharing distributions and simulation results, and the ability to share @RISK models with non-@RISK users.
Another highlight was keynote speaker David Apgar’s address presenting an interesting proposition about risk intelligence. Mr. Apgar is the author of the just-published book Risk Intelligence: Learning to Manage What We Don’t Know from Harvard Business School Press. He asserted that business risks are learnable, and not simply random, thereby allowing organizations to better manage the unknown. He provided a scorecard to help businesses quantify their risk intelligence, and suggested from there building a “risk pipeline” to constantly monitor organizational risks.
:: Diverse Delegation; Diverse Locations
A cross-section of customers participating in Palisade’s inaugural annual user conference included: S.C. Johnson & Son, Inc., Kimberly-Clark, Cummins Inc., McKinsey and Co., Capital One, WorleyParsons Komex, Boeing, General Dynamics, Banco Nacional de Costa Rica, Detroit Public Schools, Sumitomo Mitui Banking Corp. Group, United States Military Academy at West Point, ISA Colombia, Ontario Power Authority, and INCAE Business School.
The now annual Palisade User Conference: Americas was the finale in a global user conference series that began in London, England in June, and moved on to Sydney, Australia. The 2007 Palisade User Conference: Europe is already set for April 23rd and 24th in London, while dates are being planned for events in Australia, Latin America, and North America as well.
:: Presentations Available
Sessions were presented from a variety of industries on applications ranging from asset retirement to sports betting to Six Sigma manufacturing control. Many of the presentations from the Palisade User Conference: Americas, including the keynote address, are now available for free download from the Palisade web site.
Ask Amy continued from above
za/2 is the number that satisfies P(Z>za/2) = a/2, where Z follows a normal distribution with mean 0 and standard deviation 1. In words, za/2 is the z value such that the area of the right-hand tail is a/2. a/2 can be found by setting the desired confidence level equal to 100(1−a) and solving for a.
In this example (a 95% confidence level):
95 = 100(1−a)
Then a is 0.05 and a/2 is 0.025. To compute za/2 in Excel, use the NORMSINV function and enter =NORMSINV(1−a/2, 0, 1). Cell E13 of the attached example shows a Z value of approximately 1.96 for a 95% confidence interval.
To obtain an estimate for the standard deviation of the output, the @RISK statistics function RiskStdDev was placed in cell B14 and a simulation was run with just 100 iterations. This gave us a standard deviation of approximately 58.7. If we plug the above information into our formula we get
n = [ 1.96 × 58.7 / 10 ] ² = 133
Thus, at least 133 iterations should be run to be 95% sure that our estimate of the mean of the output in cell B11 is accurate within 10 units.