Models Flashcards
Practical use of a positive model?
identify mispricing of securities by recognizing patterns in actual price movement.(technical analysis)
Practical use of a normative model?
identify the potential mispricing of securities by identifying how securities should be priced (e.g. arbitrage free models).
Difference between an Abstract and an Applied model?
- Applied models are designed to address immediate real-world challenges and opportunities.
- Abstract models, also called basic models, tend to solve real-world challenges of the future.
Cross-sectional model
Cross-sectional models analyze relationships across characteristics or variables observed at a single point in time
Time-series model
Time-series models analyze behavior of a single subject or set of subjects through time.
Panel data sets
tracking multiple subjects through time and is combination of Cross-Sectional and Time-Series
Vasicek model
Vasicek’s model is a single-factor model that assumes constant volatility and that the short-term rate is mean reverting.
Cox, Ingersoll, and Ross model (CIR model)
Alters the Vasicek model to make the variance of the short-term interest rate proportional to the rate itself (no negative interest rates)
In Fixed Income what is the difference between Equilibrium models and Arbitrage-free models
Equilibrium make assumptions about the structure of the market and then use economic reasoning to model bond prices and the term structure.
Arbitrage-free models take the current interest as a given and model bond prices and the yield curve from there.
Ho and Lee model
Single-factor model that assumes that the short-term interest rate follows a normally distributed process, with a drift parameter (to match the observed curve)
Black–Derman–Toy Model (BDT model)
Interest rate model that centers on two relations: one focused on average forward rates and one on interest rate volatilities
What is the difference between a Q Measure and P Measure?
A Q-Measure is a biased measure (typicaly assumes risk neutrality) and P-Measure indicates an actual statistical probability
Examples of credit events
bankruptcy, downgrading, failure to make timely payments, certain corporate events, and government actions
Adverse selection
parties to a transaction have access to different information (information asymetry). Typically present before a transaction.
Moral Hazard
occurs when one party takes on more risk while the counter party bears the risk (after the transaction)
Three types of credit risk modeling approaches
the structural approach, the reduced-form approach, and the empirical approach
Merton Model
Structural Model whereby it assumes that default happens at the maturity of the debt if the asset value falls below the face value of the debt. (Highly restrictive model due to its assumptions)
4 important properties of the Merton Model
- Sensitivity to maturity (cum. PD increases as time increases).
- Sensitivity to asset volatility (PD increases as volatitlity increases)
- Sensitivity to leverage (PD increases as leverage increases)
- Sensitivity to Riskless rate (PD decreases as RF is higher)
What is the KMV model and what is its main output?
A structural credit risk model based on Merton that produces a credit score
Reduced-Form Models assume
Default is exogenous
Default intensity
Expected time to default (1/λ) and the probability of survival. The higher the default intensity, the shorter the expected time to default.
Jarrow–Turnbull
reduced form model that assumes that regardless of timing of default, recovery is received at the maturity date (extended version includes credit ratings)
Duffie–Singleton model
reduced form model that allows the recovery process to occur at any time and sets the recovery amount to be a fraction.
Main beliefs empirical credit model
- the default process is too difficult to be modeled
2. primary goal is to give a credit score
Altman’s Z-Score Model
Uses financial ratios to generate a Z-score which gives a relative rank of likelyhood of default. Higher score is better.
Five determinants Altman’s Z-Score
X1 = Working Capital / Total Assets X2 = Retained Earnings /Total Assets X3 = EBIT / Total Assets X4 = MVE / Book value total liabilities X5 = Sales / Total Assets
What is a factor?
factor represents a unique source of return and a unique premium in financial markets
Three major categories of factors
Macroeconomic, Fundamental, Statistical
Which two factors are added in the Fama-French Five Factor Model
Robust minus Weak (strong accounting profit)
Conservative minus Aggressive (Conservative=lower rate of investment in corporate assets)
Why are Fama-French models are of limited application to alternative asset returns that do not involve public equities?
Alternative assets do not tend to have the same factor exposures as traditional assets
What are the three observations on Factor investing from Ang (2014)
- Factors matter not assets (assets are merely a way to access factors).
- Assets are bundles of factors
- Different investors should focus on different factors
What are the 4 important practical implications of an adaptive view of the markets?
- Time-Varying Risk Premiums (risk premiums vary over time)
- Market Efficiency is a Relative Concept (it has varying degrees)
- Adaptation for Success and Survival (trading strategies should evolve with the time)
- Inevitable degredation of Alpha
Two popular approaches to modeling time-varying volatility as a stochastic process
Heston: volatility is a continuous stochastic process and volatility reverts to a long term mean.
Bates: Similar to Heston model but allows for a jump process
Explain negative volatility risk premium
Lower expected return from positive exposure to volatility because positive exposure has negative beta
An important implication of time-varying volatility
non-normality of returns
What does it mean if a portfolio dominates another portfolio
higher returns with the same level of risk
Risk averse investor
Has a concave utility function (increasing at a decreasing rate) and requires higher expected return to bear risk
What does mean-variance optimization lead to and how can this be overcome?
typically leads to unrealistic weights which can be overcome by imposing limits on the weights
Hurdle rate
hurdle rate is the minimum expected rate of return that a new asset must offer in order to be a beneficial inclusion in an otherwise existing optimal portfolio.
two types of illiquidity risks
Market liquidity risk arises when an event forces an investor to sell an asset that is not actively traded and there are a limited number of active market participants.
Funding liquidity risk arises when a borrower or investor is unable to immediately pay what is owed. A forced liquidation of assets may occur
Primary innovation of the Black–Litterman approach
it allows the investor to blend asset-specific views of each asset’s expected return
Three steps in implementing a Risk-Parity approach
- Define the total risk of the portfolio
- Define a method to measure marginal risk contribution
- Determine portfolio weights
new investment model
Beta is sought through investment products that cost-effectively offer returns driven by beta and Alpha is sought independently of beta.
