Testing and modeling causal relationships introduction
See also: Introduction to risk analysis, Types of models to analyze data
In this section we look at some practical issues of causality from a risk analysis perspective. There are a few simple, very practical and intuitive rules that will help you test a hypothesized causal relationship.
Causal inference is mostly applied to health issues, although the thinking has potential applications in other areas like econometrics, so we will use health issues as examples in this topic. We can attempt to use a causal model to answer three different types of questions:
-
Predictions - what will happen given a certain set of conditions
-
Interventions - what would be the effect of controlling one or more conditions?
-
Counterfactuals - what would have happened differently if one or more conditions had been different
In a deterministic (non-random) world there is a straightforward interpretation to causality. CSI Miami and derivatives, and all those medical dramas, are such fun programs because we viewers try to figure out what really happened - what caused this week's murder(s), and of course the program always finished with a satisfyingly unequivocal solution.
In the risk analysis world we have to work with causal relationships that are usually probabilistic in nature, for example:
-
The probability of having lung cancer within your life is x if you smoke.
-
The probability of having lung cancer within your life is y if you don't smoke.
We all know that x > y which makes being a smoker a risk factor. But life is more complicated than that: there is a biological gradient, meaning in this case the more you smoke, the more likely the cancer. If we were to do a study designed to determine the causal relationship between smoking and cancer we should look not just at whether people smoked at all, but at how much a person has smoked, for how long, and in what way (cigars, cigarettes with or without filters, pipes, little puffs or deep inhaling, brand, etc). Things are further complicated because people can change their smoking habits over time. How about:
-
The probability of having lung cancer within your life is a if you carry matches.
-
The probability of having lung cancer within your life is b if you don't carry matches.
If it would be researched, it is likely that the outcome would be a > b although carrying matches should not be a risk factor. A correct statistical analysis will determine the high correlation between carrying matches (or lighters) and using tobacco products. A sensible statistician would figure out that matches should be removed from the analysis. An uncontrolled statistical analysis can produce some silly results (imagine we had no idea that tobacco could be related to cancer and didn't collect any tobacco-related data), so we should always apply some disciplined thinking to how we structure and interpret a statistical model. We need a few definitions to begin:
A risk factor is an aspect of personal behaviour or lifestyle, environment or characteristic thought to be associated positively or negatively with a particular adverse condition.
A counterfactual world is an epidemiological hypothetical idea of a world similar to our own in all ways but for which the exposure to the hazard, or people's behaviour or characteristics, or some other change that effects exposure, have been changed in some way.
The population attributable risk (PAR) (aka population aetiological fraction, among many others) is the proportion of the incidence in the population attributable to exposure to a risk factor. It represents the fraction by which the incidence in the population would have reduced in a counterfactual world where the effect associated with that risk factor was not present.
These concepts are often used to help model what the future might look like if we were to eliminate a risk factor but we need to be careful since they technically only refer to the comparison of an observed world and a counterfactual parallel world in which the risk factor did not appear - making predictions of the future means that we have to assume that the future world would look just like that counterfactual one.
In figuring out the PAR we may well have to consider interactions between risk factors. Consider the situation where the presence of either of two risk factors give an extremely high probability of the risk of interest, and where a significant fraction of the population is exposed to both risk factors: in that case there is a lot of overlap and an individual risk factor has less impact because the other risk factor is competing for the same victims. On the other hand, exposure to two chemicals at the same time might produce a far greater effect than either chemical alone. We talk about synergism and antagonism when the risk factors work together or against each other respectively. Synergism is more common so the PAR for the combination of two or more risk factors is usually less than the sum of their individual PARs.
A food safety example: Campylobacter
A large survey conducted by CDC (the highly reputable Centers for Disease Control and Prevention) in the United States looked at why people end up getting a certain type of food poisoning (campylobacteriosis). You get campylobacteriosis when bacteria called Campylobacter enter your intestine, find a suitably protected location and multiply (form a colony). Thus, the sequence of events resulting in campylobacteriosis must include some exposure to the bacteria, then survival of those bacteria through the stomach (the acid can kill them), then setting up a colony. In order for us to observe the infection that person has to become ill. In order to identify the disease as campylobacteriosis, a doctor has to ask for a stool sample, it has to be provided, the stool sample has to be cultured and the Campylobacter isolated and identified. Campylobacteriosis will usually resolve itself after a week or so of unpleasantness so many more people therefore have campylobacteriosis than a healthcare provider will observe.
The US survey looked at behavior patterns of people with confirmed cases and tried to match them with others of the same sex, age, etc. known not to have suffered from a foodborne illness and looked for patterns of differences. This is called a case-control study. Some of the key factors were (+ meaning positively associated with illness, - meaning negatively associated):
1. Ate barbecued chicken (+);
2. Ate in a restaurant (+);
3. Were male and young (+);
4. Had healthcare insurance (+);
5. Are in a low socio-economic band (+);
6. There was another member of the family with an illness (+);
7. The person was old (+);
8. Regularly ate chicken at home (-);
9. Had a dog or cat (+); and
10. Worked on a farm (+)
Let's see whether this matches our understanding of the world:
#1 makes sense since Campylobacter naturally occurs in chicken and are very frequently to be found in chicken meat. People are also somewhat less careful with their hygiene and the cooking is less controlled at a barbecue (healthcare tip: when you've cooked a piece of meat place it on a different plate than the one used to bring the raw meat to the barbecue);
#2 makes sense because of cross-contamination in the restaurant kitchen so you might eat a veggie burger, but still have consumed Campylobacter originating from a chicken;
#3 makes sense because we guys tend not to pay much attention to kitchen practices when we're young and start off rather hopeless when we first leave home;
#4 makes sense in that, in the US, visiting a doctor is expensive and that is the only way the healthcare will observe the infection;
#5 seems right because of number 4, and maybe because poorer people will eat cheaper quality food, will visit restaurants with higher capacity and lower standards (related to number 2);
#6 is obvious since fecal-oral transmission is a known route (healthcare tip: wash your hands very well particularly when you are ill);
#7 makes sense because older people have a less robust immune system, but maybe they also eat in restaurants more (less?) often, maybe they like chicken more, etc;
#8 seems strange. It appears from a number of studies that if you eat chicken at home you are less likely to get ill. Maybe that is because it displaces eating chicken at a restaurant, maybe it's because people who cook are wealthier or care more about their food, or maybe (the current theory) it is because these people get regular small exposures to Campylobacter that boosts their immune system;
#9 is trickier. Perhaps pet food contains Campylobacter, perhaps the animal gets uncooked scraps, then cross-infects the family;
#10 makes sense. People working in chicken farms are obviously more at risk, but a farm will often have just a few chickens, or will buy in manure as fertiliser or used chicken bedding as cattle feed. Other animals also carry Campylobacter.
Each of the above is a demonstrable risk factor because each passed a test of statistical significance in this study (and others) and one can find a possible rational explanation. Of course, the possible rational explanation is often to be expected because the survey was put together with questions that were designed to test suspected risk factors, not the ones that weren't thought of. Note that the causal arguments are often inter-linked in some way making it difficult to figure out the importance of each factor in isolation. Statistical software can deal with this given the appropriate control.
Evaluating evidence - three tests
The first test of causality you should make is to consider whether there is a known or possible causal mechanism that can connect two variables together. For this, you may need to think out of the box: the history of science is full of examples where people considered something impossible despite an enormous amount of evidence to the contrary, because they were so firmly attached to their pet theory.
The second test is temporal ordering: if a change in variable A has an affect on variable B then the change in A should occur before the resultant change in B. If a person dies of radiation poisoning (B) then that person must have received a large dose of radiation (A) at some previous time. We can often test for temporal ordering with statistics, usually some form of regression.
But be careful, temporal ordering doesn't imply a causal relationship. Imagine you have a variable X that affects variables A and B, but B responds faster than A. If X is unobserved all we see is that A exhibits some behaviour that strongly correlates in some way to the previous behavior of B.
The third test is to determine in some way the size of the possible causal effect. That's where statistics comes in. From a risk analysis perspective, we are usually interested in what we can change about the world. That ultimately implies that we are only really interested in determining the magnitude of the causal relationships between variables we can control and those we are interested in. Risk analysts are not scientists - our job is not to devise new theories, but to adapt the current scientific (or financial, engineering, etc) knowledge to help decision-makers make probabilistic decisions. However, it is impossible to be able to step back and ask whether a tightly-held belief is correct, and then posing the awkward questions.
It's quite possible that we can come up with an alternative explanation of the world supported by the available evidence, which is fine, but that explanation has to be presented back to the scientific community for their blessing before we can rely on it to give decision-making advice.
Read on: Types of models to analyze data
Navigation
- Risk management
- Risk management introduction
- What are risks and opportunities?
- Planning a risk analysis
- Clearly stating risk management questions
- Evaluating risk management options
- Introduction to risk analysis
- The quality of a risk analysis
- Using risk analysis to make better decisions
- Explaining a models assumptions
- Statistical descriptions of model outputs
- Simulation Statistical Results
- Preparing a risk analysis report
- Graphical descriptions of model outputs
- Presenting and using results introduction
- Statistical descriptions of model results
- Mean deviation (MD)
- Range
- Semi-variance and semi-standard deviation
- Kurtosis (K)
- Mean
- Skewness (S)
- Conditional mean
- Custom simulation statistics table
- Mode
- Cumulative percentiles
- Median
- Relative positioning of mode median and mean
- Variance
- Standard deviation
- Inter-percentile range
- Normalized measures of spread - the CofV
- Graphical descriptionss of model results
- Showing probability ranges
- Overlaying histogram plots
- Scatter plots
- Effect of varying number of bars
- Sturges rule
- Relationship between cdf and density (histogram) plots
- Difficulty of interpreting the vertical scale
- Stochastic dominance tests
- Risk-return plots
- Second order cumulative probability plot
- Ascending and descending cumulative plots
- Tornado plot
- Box Plot
- Cumulative distribution function (cdf)
- Probability density function (pdf)
- Crude sensitivity analysis for identifying important input distributions
- Pareto Plot
- Trend plot
- Probability mass function (pmf)
- Overlaying cdf plots
- Cumulative Plot
- Simulation data table
- Statistics table
- Histogram Plot
- Spider plot
- Determining the width of histogram bars
- Plotting a variable with discrete and continuous elements
- Smoothing a histogram plot
- Risk analysis modeling techniques
- Monte Carlo simulation
- Monte Carlo simulation introduction
- Monte Carlo simulation in ModelRisk
- Filtering simulation results
- Output/Input Window
- Simulation Progress control
- Running multiple simulations
- Random number generation in ModelRisk
- Random sampling from input distributions
- How many Monte Carlo samples are enough?
- Probability distributions
- Distributions introduction
- Probability calculations in ModelRisk
- Selecting the appropriate distributions for your model
- List of distributions by category
- Distribution functions and the U parameter
- Univariate continuous distributions
- Beta distribution
- Beta Subjective distribution
- Four-parameter Beta distribution
- Bradford distribution
- Burr distribution
- Cauchy distribution
- Chi distribution
- Chi Squared distribution
- Continuous distributions introduction
- Continuous fitted distribution
- Cumulative ascending distribution
- Cumulative descending distribution
- Dagum distribution
- Erlang distribution
- Error distribution
- Error function distribution
- Exponential distribution
- Exponential family of distributions
- Extreme Value Minimum distribution
- Extreme Value Maximum distribution
- F distribution
- Fatigue Life distribution
- Gamma distribution
- Generalized Extreme Value distribution
- Generalized Logistic distribution
- Generalized Trapezoid Uniform (GTU) distribution
- Histogram distribution
- Hyperbolic-Secant distribution
- Inverse Gaussian distribution
- Johnson Bounded distribution
- Johnson Unbounded distribution
- Kernel Continuous Unbounded distribution
- Kumaraswamy distribution
- Kumaraswamy Four-parameter distribution
- Laplace distribution
- Levy distribution
- Lifetime Two-Parameter distribution
- Lifetime Three-Parameter distribution
- Lifetime Exponential distribution
- LogGamma distribution
- Logistic distribution
- LogLaplace distribution
- LogLogistic distribution
- LogLogistic Alternative parameter distribution
- LogNormal distribution
- LogNormal Alternative-parameter distribution
- LogNormal base B distribution
- LogNormal base E distribution
- LogTriangle distribution
- LogUniform distribution
- Noncentral Chi squared distribution
- Noncentral F distribution
- Normal distribution
- Normal distribution with alternative parameters
- Maxwell distribution
- Normal Mix distribution
- Relative distribution
- Ogive distribution
- Pareto (first kind) distribution
- Pareto (second kind) distribution
- Pearson Type 5 distribution
- Pearson Type 6 distribution
- Modified PERT distribution
- PERT distribution
- PERT Alternative-parameter distribution
- Reciprocal distribution
- Rayleigh distribution
- Skew Normal distribution
- Slash distribution
- SplitTriangle distribution
- Student-t distribution
- Three-parameter Student distribution
- Triangle distribution
- Triangle Alternative-parameter distribution
- Uniform distribution
- Weibull distribution
- Weibull Alternative-parameter distribution
- Three-Parameter Weibull distribution
- Univariate discrete distributions
- Discrete distributions introduction
- Bernoulli distribution
- Beta-Binomial distribution
- Beta-Geometric distribution
- Beta-Negative Binomial distribution
- Binomial distribution
- Burnt Finger Poisson distribution
- Delaporte distribution
- Discrete distribution
- Discrete Fitted distribution
- Discrete Uniform distribution
- Geometric distribution
- HypergeoM distribution
- Hypergeometric distribution
- HypergeoD distribution
- Inverse Hypergeometric distribution
- Logarithmic distribution
- Negative Binomial distribution
- Poisson distribution
- Poisson Uniform distribution
- Polya distribution
- Skellam distribution
- Step Uniform distribution
- Zero-modified counting distributions
- More on probability distributions
- Multivariate distributions
- Multivariate distributions introduction
- Dirichlet distribution
- Multinomial distribution
- Multivariate Hypergeometric distribution
- Multivariate Inverse Hypergeometric distribution type2
- Negative Multinomial distribution type 1
- Negative Multinomial distribution type 2
- Multivariate Inverse Hypergeometric distribution type1
- Multivariate Normal distribution
- More on probability distributions
- Approximating one distribution with another
- Approximations to the Inverse Hypergeometric Distribution
- Normal approximation to the Gamma Distribution
- Normal approximation to the Poisson Distribution
- Approximations to the Hypergeometric Distribution
- Stirlings formula for factorials
- Normal approximation to the Beta Distribution
- Approximation of one distribution with another
- Approximations to the Negative Binomial Distribution
- Normal approximation to the Student-t Distribution
- Approximations to the Binomial Distribution
- Normal_approximation_to_the_Binomial_distribution
- Poisson_approximation_to_the_Binomial_distribution
- Normal approximation to the Chi Squared Distribution
- Recursive formulas for discrete distributions
- Normal approximation to the Lognormal Distribution
- Normal approximations to other distributions
- Approximating one distribution with another
- Correlation modeling in risk analysis
- Common mistakes when adapting spreadsheet models for risk analysis
- More advanced risk analysis methods
- SIDs
- Modeling with objects
- ModelRisk database connectivity functions
- PK/PD modeling
- Value of information techniques
- Simulating with ordinary differential equations (ODEs)
- Optimization of stochastic models
- ModelRisk optimization extension introduction
- Optimization Settings
- Defining Simulation Requirements in an Optimization Model
- Defining Decision Constraints in an Optimization Model
- Optimization Progress control
- Defining Targets in an Optimization Model
- Defining Decision Variables in an Optimization Model
- Optimization Results
- Summing random variables
- Aggregate distributions introduction
- Aggregate modeling - Panjer's recursive method
- Adding correlation in aggregate calculations
- Sum of a random number of random variables
- Moments of an aggregate distribution
- Aggregate modeling in ModelRisk
- Aggregate modeling - Fast Fourier Transform (FFT) method
- How many random variables add up to a fixed total
- Aggregate modeling - compound Poisson approximation
- Aggregate modeling - De Pril's recursive method
- Testing and modeling causal relationships
- Stochastic time series
- Time series introduction
- Time series in ModelRisk
- Autoregressive models
- Thiel inequality coefficient
- Effect of an intervention at some uncertain point in time
- Log return of a Time Series
- Markov Chain models
- Seasonal time series
- Bounded random walk
- Time series modeling in finance
- Birth and death models
- Time series models with leading indicators
- Geometric Brownian Motion models
- Time series projection of events occurring randomly in time
- Simulation for six sigma
- ModelRisk's Six Sigma functions
- VoseSixSigmaCp
- VoseSixSigmaCpkLower
- VoseSixSigmaProbDefectShift
- VoseSixSigmaLowerBound
- VoseSixSigmaK
- VoseSixSigmaDefectShiftPPMUpper
- VoseSixSigmaDefectShiftPPMLower
- VoseSixSigmaDefectShiftPPM
- VoseSixSigmaCpm
- VoseSixSigmaSigmaLevel
- VoseSixSigmaCpkUpper
- VoseSixSigmaCpk
- VoseSixSigmaDefectPPM
- VoseSixSigmaProbDefectShiftLower
- VoseSixSigmaProbDefectShiftUpper
- VoseSixSigmaYield
- VoseSixSigmaUpperBound
- VoseSixSigmaZupper
- VoseSixSigmaZmin
- VoseSixSigmaZlower
- Modeling expert opinion
- Modeling expert opinion introduction
- Sources of error in subjective estimation
- Disaggregation
- Distributions used in modeling expert opinion
- A subjective estimate of a discrete quantity
- Incorporating differences in expert opinions
- Modeling opinion of a variable that covers several orders of magnitude
- Maximum entropy
- Probability theory and statistics
- Probability theory and statistics introduction
- Stochastic processes
- Stochastic processes introduction
- Poisson process
- Hypergeometric process
- The hypergeometric process
- Number in a sample with a particular characteristic in a hypergeometric process
- Number of hypergeometric samples to get a specific number of successes
- Number of samples taken to have an observed s in a hypergeometric process
- Estimate of population and sub-population sizes in a hypergeometric process
- The binomial process
- Renewal processes
- Mixture processes
- Martingales
- Estimating model parameters from data
- The basics
- Probability equations
- Probability theorems and useful concepts
- Probability parameters
- Probability rules and diagrams
- The definition of probability
- The basics of probability theory introduction
- Fitting probability models to data
- Fitting time series models to data
- Fitting correlation structures to data
- Fitting in ModelRisk
- Fitting probability distributions to data
- Fitting distributions to data
- Method of Moments (MoM)
- Check the quality of your data
- Kolmogorov-Smirnoff (K-S) Statistic
- Anderson-Darling (A-D) Statistic
- Goodness of fit statistics
- The Chi-Squared Goodness-of-Fit Statistic
- Determining the joint uncertainty distribution for parameters of a distribution
- Using Method of Moments with the Bootstrap
- Maximum Likelihood Estimates (MLEs)
- Fitting a distribution to truncated censored or binned data
- Critical Values and Confidence Intervals for Goodness-of-Fit Statistics
- Matching the properties of the variable and distribution
- Transforming discrete data before performing a parametric distribution fit
- Does a parametric distribution exist that is well known to fit this type of variable?
- Censored data
- Fitting a continuous non-parametric second-order distribution to data
- Goodness of Fit Plots
- Fitting a second order Normal distribution to data
- Using Goodness-of Fit Statistics to optimize Distribution Fitting
- Information criteria - SIC HQIC and AIC
- Fitting a second order parametric distribution to observed data
- Fitting a distribution for a continuous variable
- Does the random variable follow a stochastic process with a well-known model?
- Fitting a distribution for a discrete variable
- Fitting a discrete non-parametric second-order distribution to data
- Fitting a continuous non-parametric first-order distribution to data
- Fitting a first order parametric distribution to observed data
- Fitting a discrete non-parametric first-order distribution to data
- Fitting distributions to data
- Technical subjects
- Comparison of Classical and Bayesian methods
- Comparison of classic and Bayesian estimate of Normal distribution parameters
- Comparison of classic and Bayesian estimate of intensity lambda in a Poisson process
- Comparison of classic and Bayesian estimate of probability p in a binomial process
- Which technique should you use?
- Comparison of classic and Bayesian estimate of mean "time" beta in a Poisson process
- Classical statistics
- Bayesian
- Bootstrap
- The Bootstrap
- Linear regression parametric Bootstrap
- The Jackknife
- Multiple variables Bootstrap Example 2: Difference between two population means
- Linear regression non-parametric Bootstrap
- The parametric Bootstrap
- Bootstrap estimate of prevalence
- Estimating parameters for multiple variables
- Example: Parametric Bootstrap estimate of the mean of a Normal distribution with known standard deviation
- The non-parametric Bootstrap
- Example: Parametric Bootstrap estimate of mean number of calls per hour at a telephone exchange
- The Bootstrap likelihood function for Bayesian inference
- Multiple variables Bootstrap Example 1: Estimate of regression parameters
- Bayesian inference
- Uninformed priors
- Conjugate priors
- Prior distributions
- Bayesian analysis with threshold data
- Bayesian analysis example: gender of a random sample of people
- Informed prior
- Simulating a Bayesian inference calculation
- Hyperparameters
- Hyperparameter example: Micro-fractures on turbine blades
- Constructing a Bayesian inference posterior distribution in Excel
- Bayesian analysis example: Tigers in the jungle
- Markov chain Monte Carlo (MCMC) simulation
- Introduction to Bayesian inference concepts
- Bayesian estimate of the mean of a Normal distribution with known standard deviation
- Bayesian estimate of the mean of a Normal distribution with unknown standard deviation
- Determining prior distributions for correlated parameters
- Improper priors
- The Jacobian transformation
- Subjective prior based on data
- Taylor series approximation to a Bayesian posterior distribution
- Bayesian analysis example: The Monty Hall problem
- Determining prior distributions for uncorrelated parameters
- Subjective priors
- Normal approximation to the Beta posterior distribution
- Bayesian analysis example: identifying a weighted coin
- Bayesian estimate of the standard deviation of a Normal distribution with known mean
- Likelihood functions
- Bayesian estimate of the standard deviation of a Normal distribution with unknown mean
- Determining a prior distribution for a single parameter estimate
- Simulating from a constructed posterior distribution
- Bootstrap
- Comparison of Classical and Bayesian methods
- Analyzing and using data introduction
- Data Object
- Vose probability calculation
- Bayesian model averaging
- Miscellaneous
- Excel and ModelRisk model design and validation techniques
- Using range names for model clarity
- Color coding models for clarity
- Compare with known answers
- Checking units propagate correctly
- Stressing parameter values
- Model Validation and behavior introduction
- Informal auditing
- Analyzing outputs
- View random scenarios on screen and check for credibility
- Split up complex formulas (megaformulas)
- Building models that are efficient
- Comparing predictions against reality
- Numerical integration
- Comparing results of alternative models
- Building models that are easy to check and modify
- Model errors
- Model design introduction
- About array functions in Excel
- Excel and ModelRisk model design and validation techniques
- Monte Carlo simulation
- RISK ANALYSIS SOFTWARE
- Risk analysis software from Vose Software
- ModelRisk - risk modeling in Excel
- ModelRisk functions explained
- VoseCopulaOptimalFit and related functions
- VoseTimeOptimalFit and related functions
- VoseOptimalFit and related functions
- VoseXBounds
- VoseCLTSum
- VoseAggregateMoments
- VoseRawMoments
- VoseSkewness
- VoseMoments
- VoseKurtosis
- VoseAggregatePanjer
- VoseAggregateFFT
- VoseCombined
- VoseCopulaBiGumbel
- VoseCopulaBiClayton
- VoseCopulaBiNormal
- VoseCopulaBiT
- VoseKendallsTau
- VoseRiskEvent
- VoseCopulaBiFrank
- VoseCorrMatrix
- VoseRank
- VoseValidCorrmat
- VoseSpearman
- VoseCopulaData
- VoseCorrMatrixU
- VoseTimeSeasonalGBM
- VoseMarkovSample
- VoseMarkovMatrix
- VoseThielU
- VoseTimeEGARCH
- VoseTimeAPARCH
- VoseTimeARMA
- VoseTimeDeath
- VoseTimeAR1
- VoseTimeAR2
- VoseTimeARCH
- VoseTimeMA2
- VoseTimeGARCH
- VoseTimeGBMJDMR
- VoseTimePriceInflation
- VoseTimeGBMMR
- VoseTimeWageInflation
- VoseTimeLongTermInterestRate
- VoseTimeMA1
- VoseTimeGBM
- VoseTimeGBMJD
- VoseTimeShareYields
- VoseTimeYule
- VoseTimeShortTermInterestRate
- VoseDominance
- VoseLargest
- VoseSmallest
- VoseShift
- VoseStopSum
- VoseEigenValues
- VosePrincipleEsscher
- VoseAggregateMultiFFT
- VosePrincipleEV
- VoseCopulaMultiNormal
- VoseRunoff
- VosePrincipleRA
- VoseSumProduct
- VosePrincipleStdev
- VosePoissonLambda
- VoseBinomialP
- VosePBounds
- VoseAIC
- VoseHQIC
- VoseSIC
- VoseOgive1
- VoseFrequency
- VoseOgive2
- VoseNBootStdev
- VoseNBoot
- VoseSimulate
- VoseNBootPaired
- VoseAggregateMC
- VoseMean
- VoseStDev
- VoseAggregateMultiMoments
- VoseDeduct
- VoseExpression
- VoseLargestSet
- VoseKthSmallest
- VoseSmallestSet
- VoseKthLargest
- VoseNBootCofV
- VoseNBootPercentile
- VoseExtremeRange
- VoseNBootKurt
- VoseCopulaMultiClayton
- VoseNBootMean
- VoseTangentPortfolio
- VoseNBootVariance
- VoseNBootSkewness
- VoseIntegrate
- VoseInterpolate
- VoseCopulaMultiGumbel
- VoseCopulaMultiT
- VoseAggregateMultiMC
- VoseCopulaMultiFrank
- VoseTimeMultiMA1
- VoseTimeMultiMA2
- VoseTimeMultiGBM
- VoseTimeMultBEKK
- VoseAggregateDePril
- VoseTimeMultiAR1
- VoseTimeWilkie
- VoseTimeDividends
- VoseTimeMultiAR2
- VoseRuinFlag
- VoseRuinTime
- VoseDepletionShortfall
- VoseDepletion
- VoseDepletionFlag
- VoseDepletionTime
- VosejProduct
- VoseCholesky
- VoseTimeSimulate
- VoseNBootSeries
- VosejkProduct
- VoseRuinSeverity
- VoseRuin
- VosejkSum
- VoseTimeDividendsA
- VoseRuinNPV
- VoseTruncData
- VoseSample
- VoseIdentity
- VoseCopulaSimulate
- VoseSortA
- VoseFrequencyCumulA
- VoseAggregateDeduct
- VoseMeanExcessP
- VoseProb10
- VoseSpearmanU
- VoseSortD
- VoseFrequencyCumulD
- VoseRuinMaxSeverity
- VoseMeanExcessX
- VoseRawMoment3
- VosejSum
- VoseRawMoment4
- VoseNBootMoments
- VoseVariance
- VoseTimeShortTermInterestRateA
- VoseTimeLongTermInterestRateA
- VoseProb
- VoseDescription
- VoseCofV
- VoseAggregateProduct
- VoseEigenVectors
- VoseTimeWageInflationA
- VoseRawMoment1
- VosejSumInf
- VoseRawMoment2
- VoseShuffle
- VoseRollingStats
- VoseSplice
- VoseTSEmpiricalFit
- VoseTimeShareYieldsA
- VoseParameters
- VoseAggregateTranche
- VoseCovToCorr
- VoseCorrToCov
- VoseLLH
- VoseTimeSMEThreePoint
- VoseDataObject
- VoseCopulaDataSeries
- VoseDataRow
- VoseDataMin
- VoseDataMax
- VoseTimeSME2Perc
- VoseTimeSMEUniform
- VoseTimeSMESaturation
- VoseOutput
- VoseInput
- VoseTimeSMEPoisson
- VoseTimeBMAObject
- VoseBMAObject
- VoseBMAProb10
- VoseBMAProb
- VoseCopulaBMA
- VoseCopulaBMAObject
- VoseTimeEmpiricalFit
- VoseTimeBMA
- VoseBMA
- VoseSimKurtosis
- VoseOptConstraintMin
- VoseSimProbability
- VoseCurrentSample
- VoseCurrentSim
- VoseLibAssumption
- VoseLibReference
- VoseSimMoments
- VoseOptConstraintMax
- VoseSimMean
- VoseOptDecisionContinuous
- VoseOptRequirementEquals
- VoseOptRequirementMax
- VoseOptRequirementMin
- VoseOptTargetMinimize
- VoseOptConstraintEquals
- VoseSimVariance
- VoseSimSkewness
- VoseSimTable
- VoseSimCofV
- VoseSimPercentile
- VoseSimStDev
- VoseOptTargetValue
- VoseOptTargetMaximize
- VoseOptDecisionDiscrete
- VoseSimMSE
- VoseMin
- VoseMin
- VoseOptDecisionList
- VoseOptDecisionBoolean
- VoseOptRequirementBetween
- VoseOptConstraintBetween
- VoseSimMax
- VoseSimSemiVariance
- VoseSimSemiStdev
- VoseSimMeanDeviation
- VoseSimMin
- VoseSimCVARp
- VoseSimCVARx
- VoseSimCorrelation
- VoseSimCorrelationMatrix
- VoseOptConstraintString
- VoseOptCVARx
- VoseOptCVARp
- VoseOptPercentile
- VoseSimValue
- VoseSimStop
- Precision Control Functions
- VoseAggregateDiscrete
- VoseTimeMultiGARCH
- VoseTimeGBMVR
- VoseTimeGBMAJ
- VoseTimeGBMAJVR
- VoseSID
- Generalized Pareto Distribution (GPD)
- Generalized Pareto Distribution (GPD) Equations
- Three-Point Estimate Distribution
- Three-Point Estimate Distribution Equations
- VoseCalibrate
- ModelRisk interfaces
- Integrate
- Data Viewer
- Stochastic Dominance
- Library
- Correlation Matrix
- Portfolio Optimization Model
- Common elements of ModelRisk interfaces
- Risk Event
- Extreme Values
- Select Distribution
- Combined Distribution
- Aggregate Panjer
- Interpolate
- View Function
- Find Function
- Deduct
- Ogive
- AtRISK model converter
- Aggregate Multi FFT
- Stop Sum
- Crystal Ball model converter
- Aggregate Monte Carlo
- Splicing Distributions
- Subject Matter Expert (SME) Time Series Forecasts
- Aggregate Multivariate Monte Carlo
- Ordinary Differential Equation tool
- Aggregate FFT
- More on Conversion
- Multivariate Copula
- Bivariate Copula
- Univariate Time Series
- Modeling expert opinion in ModelRisk
- Multivariate Time Series
- Sum Product
- Aggregate DePril
- Aggregate Discrete
- Expert
- ModelRisk introduction
- Building and running a simple example model
- Distributions in ModelRisk
- List of all ModelRisk functions
- Custom applications and macros
- ModelRisk functions explained
- Tamara - project risk analysis
- Introduction to Tamara project risk analysis software
- Launching Tamara
- Importing a schedule
- Assigning uncertainty to the amount of work in the project
- Assigning uncertainty to productivity levels in the project
- Adding risk events to the project schedule
- Adding cost uncertainty to the project schedule
- Saving the Tamara model
- Running a Monte Carlo simulation in Tamara
- Reviewing the simulation results in Tamara
- Using Tamara results for cost and financial risk analysis
- Creating, updating and distributing a Tamara report
- Tips for creating a schedule model suitable for Monte Carlo simulation
- Random number generator and sampling algorithms used in Tamara
- Probability distributions used in Tamara
- Correlation with project schedule risk analysis
- Pelican - enterprise risk management