1. Common probability distributions include the discrete uniform, binomial, normal (Gaussian), and lognormal distributions. The normal distribution is a symmetrical bell-shaped curve used to represent variables and the lognormal distribution models values that are always positive.
2. Monte Carlo simulation is a technique used to understand the impact of risk and uncertainty. It involves generating thousands of possible outcomes based on probability distributions and risk factors to estimate outcomes of a project or investment.
3. Historical simulation is another forecasting technique that uses actual historical changes in values or risk factors over a prior period to simulate possible future outcomes without estimating distributions. While it avoids estimation, future outcomes may fall outside the historical range.
1. Common probability distributions include the discrete uniform, binomial, normal (Gaussian), and lognormal distributions. The normal distribution is a symmetrical bell-shaped curve used to represent variables and the lognormal distribution models values that are always positive.
2. Monte Carlo simulation is a technique used to understand the impact of risk and uncertainty. It involves generating thousands of possible outcomes based on probability distributions and risk factors to estimate outcomes of a project or investment.
3. Historical simulation is another forecasting technique that uses actual historical changes in values or risk factors over a prior period to simulate possible future outcomes without estimating distributions. While it avoids estimation, future outcomes may fall outside the historical range.
1. Common probability distributions include the discrete uniform, binomial, normal (Gaussian), and lognormal distributions. The normal distribution is a symmetrical bell-shaped curve used to represent variables and the lognormal distribution models values that are always positive.
2. Monte Carlo simulation is a technique used to understand the impact of risk and uncertainty. It involves generating thousands of possible outcomes based on probability distributions and risk factors to estimate outcomes of a project or investment.
3. Historical simulation is another forecasting technique that uses actual historical changes in values or risk factors over a prior period to simulate possible future outcomes without estimating distributions. While it avoids estimation, future outcomes may fall outside the historical range.
1. Probability - All possible outcomes for a variable 14. Continuous
Distribution - Probabilities of all possible outcomes = 1 Uniform Distribution 2. Discrete - Number of possible outcomes that can be Formula Random counted Variable - Each outcomes there is a measurable and 15. Normal - Function that represents the distribution of positive probability Distribution variables as a symmetrical bell-shaped 3. Probability - Denoted p(x) graph Function - Random variable is equal to specific value - Skewness = 0 - Kurtosis = 3 4. Key 0 < p(x) < 1 Properties of ∑p(x) = 1 16. Multivariate - Specifies the probabilities for a group of Probability Distribution related random variables Function - meaningful only if those variables are dependent on each other 5. Continuous - the number of possible outcomes is infinite, Random even if lower and upper bounds exist 17. Confidence the range of values within which a Variable - time is a variable Interval population parameter is estimated to lie 6. Discrete p(x) = 0 when x cannot occur 18. Three Popular 90% confidence interval = mean of - 1.65 to Distribution p(x) > 0 when x can occur Confidence +1.65 Intervals 95% confidence interval = mean of - 1.96 to 7. Continuous p(x) = 0 when x can occur +1.96 Distribution p(x₁ < x < x₂) where x₁/x₂ are actual numbers 99% confidence interval = mean of - 2.58 to 8. Cumulative - Another name is "distribution function" +2.58 Distribution - Defines the probability that a random 19. Standard A normal distribution with a mean of 0 and Function variable takes on a value equal to or less Normal a standard deviation of 1 (cdf) than a specific Distribution - Typically F(X) = P(X) plus everything before that 20. Z-Value Number of standard errors that a point is away from the mean 9. Discrete probabilities for all possible outcomes for a Uniform discrete random variable are equal. 21. Z-Value Forula Random Variable 10. Binomial - The number of "successes" in a given 22. Roy's Safety- The optimal portfolio minimizes the Random number of trials, whereby the outcome can First Ratio probability that the return of the portfolio Variable be either "success" or "failure" falls below a minimum acceptable level; - Only 2 trials = (Historical Return - Return Threshold)/(Volatility) 11. Binomial Probability 23. SF Ratio Formula Formula
12. Expected 24. Lognormal - Function e^x where x is normally
Value of a Distribution distributed; Binomial - Positively skewed; Probability - Bound to the left by 0 - if x is normal, e^x is lognormal 25. Discretely Compound returns we are familiar with Compounded given some discrete compounding period Returns - Same as EAR (effective annual rate) 13. Continuous defined over a range that spans between Uniform some lower limit, a, and some upper limit b, Distribution which serves as the parameters of the distribution R9 Common Probability Distributions 02
26. Continuous - Shorter and shorter periods of compounding
Compounding - Formula: e^r - 1 - Formula: ln(S1/S2) - Formula: ln(1 + HPR) 27. Monte Carlo A process which generates hundreds or thousands of probable performance outcomes based on probability Simulation distributions for cost and schedule on individual tasks. The outcomes are then used to generate a probability distribution for the project as a whole 28. Historical - Based on actual changes in value or actual changes in risk factors over some prior period Simulation - Pros: Don't have to estimate distribution of risk factors - Cons: Future outcomes for risk factors may be outside the historical range