Introduction to @RISK Software Monte Carlo Simulation Analysis
This article is translated by China Science Software Network
@RISK software Monte Carlo simulation analysis
Risk analysis is part of every decision we make. We often face uncertainty, ambiguity and variability. Moreover, even if we can make unprecedented access to information, we still can't accurately predict the future. Monte Carlo simulation allows you to view all possible outcomes of your decisions and assess the impact of the risk, making better decisions in the presence of uncertainties.
What is Monte Carlo simulation?
Monte Carlo simulation is a computerized mathematical method that allows people to assess the risks involved in quantitative analysis and decision making. Professionals apply this method across a wide range of areas such as finance, project management, energy, manufacturing, engineering, research and development, insurance, oil and gas, transportation and the environment.
The Monte Carlo simulation provides decision makers with a range of possible outcomes and probabilities that may result from any action taken. It illustrates the greatest possibilities, the result of all-out and most conservative decisions, and all possible consequences of compromise decisions.
This method was first used by scientists to study the atomic bomb; it was named after Monte Carlo, the tourist city of Monaco, which is famous for its casinos. Since its introduction in World War II, Monte Carlo simulations have been used to model different physical and conceptual systems.
How Monte Carlo simulation works
Monte Carlo simulation performs a risk analysis by substituting a range of values ​​(probability distribution) for a model with possible outcomes for any factor with inherent uncertainty. The results are then repeated, each time using a different set of random values ​​in the distribution function. Based on the number and range of uncertainties specified for random values, Monte Carlo simulations can be recalculated thousands of times before completion. Monte Carlo simulations generate a distribution of possible outcome values.
By using a probability distribution, the variables can have different probabilities of the different outcomes generated. Probability distribution is a more realistic approach to describing uncertainties in variables of risk analysis. Common probability distributions include:
Normal distribution - or "bell curve". The user only needs to define an average or expected value and standard deviation to describe the change in the mean. The median value near the mean is most likely to occur. This distribution is symmetrical and describes many natural phenomena, such as the height of a person. Examples of variables described by a normal distribution include inflation rate and energy price.
Lognormal distribution—values ​​are positively skewed, rather than symmetric distributions of normal distributions. It is used to represent a value that is not below zero but has an infinite positive probability. Examples of variables described by the lognormal distribution include real estate value, stock price, and oil reserves.
Uniform distribution - all values ​​have the same chance of occurrence, and the user only needs to define the minimum and maximum values. Examples of variables that can be evenly distributed include manufacturing costs or future sales revenue for new products.
Triangle distribution — User-defined minimum, most likely, and maximum. The values ​​around the most likely values ​​are most likely to occur. Variables that can be described by a triangular distribution include past sales history and inventory levels per unit of time.
PERT distribution — User-defined minimum, most likely, and maximum, as with a triangular distribution. The values ​​around the most likely values ​​are most likely to occur. However, the value between the most likely value and the extreme value is more likely to occur than the triangular distribution; that is, the extreme value is no longer emphasized. An example of using a PERT distribution is to describe the duration of a task in the project management model.
Discrete Distribution—User-defined specific values ​​that may occur and the likelihood that each value will occur. An example could be litigation outcome: a 20% probability is a positive ruling, a 30% probability is a negative ruling, a 40% probability is a reconciliation, and a 10% probability is a null trial.
During the Monte Carlo simulation, the values ​​are randomly sampled from the input term probability distribution. Each set of samples is called an iteration and the results generated by the samples are recorded. Monte Carlo simulations perform hundreds or thousands of such operations, and the result is a probability distribution of possible outcomes. In this way, Monte Carlo simulations provide a more comprehensive view of what might happen. It not only tells you what is likely to happen, but also tells you the probability of the outcome.
Compared to deterministic analysis or "single point estimation" analysis, Monte Carlo simulation offers many advantages:
Probability results. The results not only show what might happen, but also show the probability of occurrence of each result.
Graphical results. Using data generated by Monte Carlo simulation, users can easily create graphs of different outcomes and their probability of occurrence. This is important for communicating results to other stakeholders.
sensitivity analysis. In only a few cases, deterministic analysis is difficult to see which variables have the greatest impact on the results. In Monte Carlo simulation, users can easily see which inputs have the greatest impact on the final result.
Scenario analysis: In deterministic models, it is difficult to model different combinations of values ​​for different inputs to see the impact of truly different scenarios. Using Monte Carlo simulations, analysts can see exactly which inputs combine which values ​​together when certain results occur. This is very important to complete further analysis.
Input relevance. In Monte Carlo simulation, models can be modeled for interdependencies between input variables. In fact, when some factors rise and other factors rise or fall accordingly, it is important to accurately indicate how they change.
The enhancement of the Monte Carlo simulation is a sampling using the Latin hypercube method, which performs a more accurate sampling from the entire range of the distribution function.
Palisade's Monte Carlo simulation products
The spreadsheet application for personal computers provides an opportunity for professionals to use Monte Carlo simulation in their daily analytical work. Microsoft Excel is the primary spreadsheet analysis tool, and Palisade's @RISK is the premier Monte Carlo simulation plug-in for Excel. In 1987, @RISK, first introduced for Lotus 1-2-3 for DOS, has a good reputation for computational accuracy, model building flexibility, and ease of use. The introduction of Microsoft Project led to another logical application of Monte Carlo simulation, which analyzes the uncertainties and risks in large project management.
pet bird toys,pet mouse toys,interactive pet toys,pet bunny toys
Ningbo XISXI E-commerce Co., Ltd , https://www.petspetstoys.com