Elliot Book Chapter 10: Risk Modeling Flashcards

Vocabulary-style flashcards covering key terms and concepts from Chapter 10 regarding methods and limitations of risk modeling, event consequences, and portfolio theory.

Empirical probability

A probability measure also known as a posteriori probability that is based on actual experience through historical data or from the observation of facts.

Theoretical probability

A probability that is based on theoretical principles rather than on actual experience, often used when historical data points are insufficient.

Normal distribution

A probability distribution that, when graphed, generates a bell-shaped curve where the majority of results fall within expected parameters.

Extreme value theory (EVT)

A statistical probability estimation of extreme deviations from the median that addresses the tail of probability distributions to forecast unlikely but possible downside risks.

Regression analysis

A statistical technique used to estimate relationships between variables, assuming the dependent variable being forecast varies predictably with an independent variable.

Preference among bets

A risk modeling technique that converts expert opinion into probabilities by presenting an expert with the choice of sides on a bet regarding certain events.

Political risk

Any action by a government that favors domestic over foreign organizations or poses a threat to foreign organizations.

Judgments of relative likelihood

A representation of probability using expert input regarding the likelihood of event outcomes relative to known probabilities, such as more, less, or equally likely.

Delphi technique

A collaborative estimating strategy using expert input to reach consensus by continuously refining individual responses through multiple rounds of questionnaires.

Monte Carlo simulation

A computerized mathematical technique that randomly selects values for variables according to a probability distribution to generate thousands of possible scenarios and outcomes.

Fuzzy logic

A type of logic that assigns values to indefinite data fields based on degrees of truth rather than absolute true/false values to facilitate more accurate probability.

Decision tree analysis

An analysis that examines the consequences, including costs and gains, of decisions to compare alternative strategies and select the best course of action.

Event tree analysis

An analysis that examines all possible consequences of an accidental event, their probabilities, and existing measures to prevent or control them.

Accidental event

Defined as the first significant deviation from a normal situation that may lead to unwanted consequences.

Influence diagram

A graphical representation consisting of nodes and arrows that provides a visual graph of a decision, its influences, and uncertainties.

Decision nodes

Elements in an influence diagram represented by rectangles.

Variables

Elements in an influence diagram represented by ovals that indicate uncertainties affecting the outcome.

Benefits and costs

Elements in an influence diagram represented by diamonds indicating outcomes.

Bayesian network

A type of influence diagram that includes probability tables in addition to a graph for mathematical probabilities of uncertainties.

Risk dashboard

A computer interface that reports quantitative data regarding an organization's key risk indicators.

Key risk indicator (KRI)

A tool that an organization uses to measure the uncertainty of meeting an organizational objective.

Correlation

A statistical relationship between variables expressed as a number from $-1$ to $+1$.

Covariance

The relative association between variables to move in tandem or independently of each other.

Correlation coefficient

A scaled version of covariance expressed as a value between $-1$ and $+1$ indicating the degree and direction of relationship.

Causality

The relationship between two events where the first event brings about the second, defining cause and effect.

Correlation matrix

A report showing the correlation or covariance for pairs of risk sources, always featuring a value of $+1$ along the diagonal.

Modern Portfolio Theory (MPT)

A theory stating that investors can optimize risk and return through diversification by considering how portfolio risks and returns interact.

Efficient frontier

A line on a graph representing the set of available investment portfolios for which risk and return are optimized to the highest expected return for a given level of risk.

Value at risk (VaR)

A threshold value such that the probability of loss on a portfolio over a given time horizon exceeds this value, assuming normal markets and no trading.

Earnings at risk (EaR)

A financial measure representing the maximum expected loss of earnings within a specific degree of confidence, such as $95\%$, often modeled using Monte Carlo simulations.

Conditional value at risk (CVaR)

A model used to determine the likelihood of a loss given that the loss is already greater than or equal to the VaR threshold.