Articles

Exam FRM Level 1 Ultimate Guide (Book 4)

Estimated reading time: 29 minutes

 

Introduction

We now turn our attention to book 4 in the FRM Level 1 course.

(click here for the previous guide)

You may download this content on our website’s shop page, free of charge.

 

FRM Level 1 Book 4 – Valuation and Risk Models

 

Measures of Financial Risk

Consider the two risk measures: value-at-risk (VaR) and expected shortfall (ES)

The mean and standard deviation of a portfolio’s return can be calculated using the means and standard deviations of its components’ returns

This leads to an important concept known as the efficient frontier

The efficient frontier shows the trade-offs between mean and standard deviation that are available to the holder of a well-diversified portfolio

When portfolios are described by the mean and standard deviation of their returns, it is natural to assume we are dealing with normal distributions

Most financial variables have fatter tails than the normal distribution

In other words, extreme events are more likely to occur than the normal distribution would predict

 

Value-at-risk (VaR)

Value-at-risk (VaR) and expected shortfall are two risk measures focusing on adverse events

Expected shortfall is less intuitive than VaR

However, expected shortfall has more desirable theoretical properties

Expected shortfall is also an example of a coherent risk measure

Investors are typically faced with a trade-off between risk and return

The greater the risks that are taken, the higher the expected return that can be achieved

Expected return does not describe the return that we expect to happen

The term is actually used by statisticians to describe the average (or mean) return

 

Normal Distribution

The normal (or Gaussian) distribution is one of the most well-known and widely used probability distributions

The normal (or Gaussian) distribution has two parameters: The mean and the standard deviation

Value at Risk, VaR, is an important risk measure that focuses on adverse events and their probability

The VaR for an investment opportunity is a function of two parameters: The time horizon, and the confidence level

One problem with VaR is that it does not say how bad losses might be when they exceed the VaR level

Expected shortfall is one way of overcoming this disadvantage

Expected shortfall is defined as the expected (or average) loss conditional on the loss being greater than the VaR level

 

Being Coherent

A risk measure that satisfies all four conditions below is termed coherent:

Monotonicity: If a portfolio always produces a worse result than another portfolio, it should have a higher risk measure

Translation Invariance: If an amount of cash K is added to a portfolio, its risk measure should decrease by K

Homogeneity: Changing the size of a portfolio by multiplying the amounts of all the components by λ results in the risk measure being multiplied by λ

Subadditivity: For any two portfolios, A and B, the risk measure for the portfolio formed by merging A and B should be no greater than the sum of the risk measures for portfolios A and B

As a risk measure, Standard Deviation is a good way of describing the overall uncertainty associated with a set of outcomes

 

Calculating and Applying VaR

Previously we looked at two risk measures: value-at-risk (VaR) and expected shortfall (ES). This segment discusses how they can be calculated:

One popular approach is a non-parametric method, where the future behavior of the underlying market variables is determined in a very direct way from their past behavior

This method is known as historical simulation

If we assume the returns on the underlying variables are multivariate normal, then:

  • Changes in portfolio value are also normally distributed
  • Calculating VaR will then be relatively straightforward

For a portfolio that is not linearly dependent on the underlying market variables (e.g., because it contains options), the delta-normal model can also be used. (However, it is less accurate in this context)

The delta-normal model works well for linear portfolios when the risk factor probability distributions are at least approximately normal

 

Monte Carlo Simulations

An alternative to historical simulation and the delta-normal model is provided by Monte Carlo simulations

These are like historical simulation, but their scenarios are randomly generated (rather than being determined directly from the behavior of market variables in the past)

Monte Carlo simulations generate scenarios by taking random samples from the distributions assumed for the risk factors

 

Historical simulation

Historical simulation is a popular method of calculating VaR and ES

Typically, the time horizon is chosen as one day

Historical simulation involves identifying the market variables on which the value of the portfolio under consideration depends

In practice, risk factors are divided into two categories:

  • Those where the percentage change in the past is used to define a percentage change in the future, and
  • Those where the actual change in the past is used to define an actual change in the future

Standard deviations increase during stressed market conditions (It is also true that correlations generally increase during this period)

 

Measuring and Monitoring Volatility

The volatility of a variable measures the extent to which its value changes through time

A constant volatility would be fairly easy to estimate using historical data

In practice, however, volatility changes through time

This leads to situations where asset returns are not normally distributed; instead, they tend to have fatter tails than a normal distribution would predict

This is important for the estimation of risk measures (such as VaR and expected shortfall) because these measures depend critically on the tails of asset return distributions

 

The conditionally normal model

An alternative to assuming asset returns is constantly normal is to assume they are normal conditioned on the volatility being known

When volatility is high, the daily return is normal with a high standard deviation

When the volatility is low, the daily return is normal with a low standard deviation

The conditionally normal model may not be perfect, but it is an improvement over the constant volatility model

To implement the conditionally normal model, it is necessary to monitor volatility so that a current volatility estimate is produced. We consider two ways of doing that:

  • The exponentially weighted moving average (EWMA) model and
  • The GARCH (1,1) model

There are three ways in which an asset’s return can deviate from normality:

  • The return distribution can have fatter tails than a normal distribution
  • The return distribution can be non-symmetrical
  • The return distribution can be unstable with parameters that vary through time

 

Standard Error of an Estimate

The standard error of an estimate is the standard deviation of the difference between the estimate and the true value

The standard error of a volatility estimate calculated from m observations is approximately equal to the estimate divided by the square root of 2(m – 1)

One way of overcoming problems with estimating volatility is to use exponential smoothing

This is also referred to as the exponentially weighted moving average (EWMA)

In EWMA, the weights applied to historical data decline exponentially as we move back in time

An alternative method is the multivariate density estimation (MDE)

The GARCH model, developed by Robert Engel and Tim Bollerslev, can be regarded as an extension of EWMA

 

Implied volatility

Implied volatility is the volatility implied from options prices

  • It is a forward-looking estimate of volatility
  • It is often found to be more accurate than an estimate produced from historical data
  • As volatilities increase, the value of the option increases

The correlation between two variables is their covariance divided by the product of their standard deviations

 

External and Internal Credit Ratings

Rating agencies are an important external source of credit risk data

The most well-known credit rating agencies are: Moody’s, Standard and Poor’s (S&P) and Fitch

The Dodd-Frank Act now requires rating agencies to make the assumptions and methodologies underlying their ratings more transparent

It has also increased the potential legal liability of rating agencies

An external credit rating is usually an attribute of an instrument issued by an entity (rather than of the entity itself)

However, bond ratings are often assumed to be attributes of the entity rather than of the bond itself

 

Rating levels

The highest bond rating assigned by Moody’s is Aaa

Bonds with an Aaa rating are considered to have almost no chance of defaulting

The next highest rating is Aa

Rating agencies rate publicly traded bonds and money market instruments

In addition to the ratings themselves, rating agencies provide what are termed outlooks

 

Rating outlooks

Outlooks are indications of the most likely direction of the rating over the medium term

  • A positive outlook means that a rating may be raised
  • A negative outlook means that it may be lowered
  • A stable outlook means it is not likely to change

Banks and other financial institutions develop their own internal rating systems based on their assessment of potential borrowers

Banks and other financial institutions typically base their ratings on several factors: Financial ratios, Cash flow projections, An assessment of the firm’s management, etc.

 

Country Risk

Many large firms have business interests all over the world

For such global entities, it is important to assess the risks associated with the foreign countries they operate in:

These risks are collectively referred to as country risk

There are numerous components of country risk. One such component is political risk

Individuals and corporations can obtain diversification benefits by investing outside their domestic markets

When lending to foreign governments, it is important for lenders to consider country risk as part of their credit default risk framework

Many developing markets have economies that are growing faster than those of developed markets

However, this fast growth may be accompanied by higher economic risks and less stable political climates

There are often links between political and economic risks

 

Gross Domestic Product (GDP)

The growth of a country’s economy is measured its Gross Domestic Product (GDP)

GDP is the total value of goods and services produced by all the people and firms in a country

An important consideration in assessing country risk is how a country will react to economic cycles

  • During economic down-turns, developing countries often see larger declines in GDP than their developed counterparts
  • This is because developing economies tend to rely more heavily on commodities

Some countries are highly dependent on a single commodity

If the price of that commodity declines, the country and the value of its currency will suffer

Many African and Latin American countries fall into this category

 

Legal Risk

Legal risk is the risk of losses due to inadequacies or biases in a country’s legal system

A legal system that is trusted and perceived to be fair helps a country to attract foreign investment

One measure of a country’s risk is the risk it will default on its debt

There are two types of sovereign debt:

  1. The type issued in a foreign currency (such as the USD)
  2. The type issued in the country’s own currency

Debt issued in a foreign currency is attractive to global banks and other international lenders

The risk for the issuing country is that it cannot repay the debt by simply printing more money

There are several other factors that are considered when determining a rating: Social Security Commitments, The Tax Base, Political Risk, Implicit Guarantees etc.

 

Credit spread

The credit spread for sovereign debt in a specific currency is the excess interest paid over the risk-free rate in that currency

There is a strong correlation between credit spreads and ratings

Credit spreads can provide extra information on the ability of a country to repay its debt

 

Measuring Credit Risk

Economic capital is a bank’s own estimate of the capital it requires

Regulatory capital is the capital bank regulators (also known as bank supervisors) require a bank to keep

Global bank regulatory requirements are determined by the Basel Committee on Banking Supervision in Switzerland

The three different models for quantifying credit risk in this segment:

The first is a model where the mean and standard deviation of the loss from a loan portfolio is determined from the properties of the individual loans

The second model, known as the Vasicek model, is used by bank regulators to estimate an extreme percentile of the loss distribution

The third model is known as CreditMetrics and is often used by banks themselves when estimating economic capital

Typically, CreditMetrics involves time-consuming Monte Carlo simulations

A bank must keep capital for the risks it takes

When losses are incurred, they come out of the equity capital

 

Equity capital and Debt capital

Equity capital is sometimes referred to as “going concern capital” since as it is positive, the bank is solvent and can therefore be characterized as a going concern

Debt capital is referred to as “gone concern capital” since it only becomes an important cushion for depositors when the bank is no longer a going concern (i.e., insolvent)

Banks are subject to many risks

Credit risk has traditionally been the most important risk taken by banks

Each loan issued by a bank has some risk of default and therefore a risk of a credit loss

It is the possibility of this loss that constitutes credit risk

In 1974, the central banks of the G10 countries formed the Basel Committee to harmonize global bank regulation

By 1988, the committee had agreed on a common approach for determining the required credit risk capital for the banks under their supervision (This regulation is now known as Basel I)

Both regulatory capital and economic capital feature separate capital calculations for credit risk, market risk, and operational risk

For regulatory capital, the results are added to give the total capital requirements

For economic capital, however, correlations between the risks are often considered

 

The Vasicek Model

The Vasicek model is used by regulators to determine capital for loan portfolios

It uses the Gaussian copula model to define the correlation between defaults

The Vasicek model has an advantage in that the unexpected loss can be determined analytically

 

CreditMetrics

CreditMetrics is the model banks often use to determine economic capital

Under this model, each borrower is assigned an external or internal credit rating

A one-year transition table is used to define how ratings change

 

Operational Risk

Operational risk is sometimes defined very broadly as any risk that is not a market risk or a credit risk

A much narrower definition would be that it consists of risks arising from operational mistakes;

This would include the risk that a bank transaction is processed incorrectly

But it would not include the risk of fraud, cyberattacks, or damage to physical assets

Operational risk has been defined by the Basel Committee as:

The risk of loss resulting from inadequate or failed internal processes, people, and systems or from external events

 

Categories of operational risk

Seven categories of operational risk have been identified by the Basel Committee

  1. Internal fraud
  2. External fraud
  3. Employment practices and work place safety
  4. Clients, products, and business practices
  5. Damage to physical assets
  6. Business disruption and system failures
  7. Execution, delivery, and process management

 

Compliance risk

Compliance risk is another operational risk facing financial institutions

This is the risk that an organization will incur fines or other penalties because it knowingly or unknowingly fails to act in accordance with industry laws and regulations, internal policies, or prescribed best practices

This includes activities such as money laundering, terrorism financing, and assisting clients with tax evasion

 

Rogue trader risk

Rogue trader risk is the risk that an employee will take unauthorized actions resulting in large losses

One of the most notorious incidents involved Barings Bank trader Nick Leeson

 

Basel II rules

The final Basel II rules for banks had three approaches:

  1. The basic indicator approach
  2. The standardized approach
  3. The advanced measurement approach (AMA)

The key determinants of an operational risk loss distribution are:

  • Average Loss frequency: the average number of times in a year that large losses occur, and
  • Loss severity: the probability distribution of the size of each loss

 

Stress Testing

Stress testing is a risk management activity that has become increasingly important since the 2007–2008 financial crisis:

  • It involves evaluating the implications of extreme scenarios that are unlikely and yet plausible
  • It asks if a financial institution has enough capital / liquid assets to survive various scenarios
  • Some stress tests are carried out because they are required by regulators
  • Others are carried out as part of their internal risk management activities

 

Stress Testing Measures

Measures such as value-at-risk (VaR) and expected shortfall (ES) are often calculated and used in stress testing analysis

One disadvantage of VaR and ES is that they are usually backward-looking

They assume the future will (in some sense) be like the past

Stress testing, however, is designed to be forward-looking

Using stress tests to derive the full range of all possible outcomes is not usually possible

Risk managers have two types of analyses available to them:

  • One is a backward-looking analysis where a loss distribution can be estimated
  • The other is a forward-looking analysis where different scenarios are assessed

In the case of market risk, the VaR/ES approach often has a short time horizon (perhaps only one day), whereas stress testing usually looks at a much longer period

 

Choosing a stress-test scenario

The first step in choosing a stress-test scenario is to select a time horizon

The time horizon should be long enough for the full impact of the scenarios to be evaluated

Scenarios lasting three months to two years are more common

Very long scenarios can be necessary in some situations

Stress testing involves constructing scenarios and then evaluating their consequences

Reverse stress testing takes the opposite approach: It asks the question, “What combination of circumstances could lead to the failure of the financial institution?”

A financial institution should have written policies and procedures for stress testing and ensure that they are adhered to

 

Pricing Conventions, Discounting, and Arbitrage

Discount factors are numbers that allow us to relate a cash flow received in the future to its value today

Treasury bills are instruments issued by a government to finance its short-term funding needs. They last one year or less and are defined by: Their face value (or principal amount or par value), and their maturity date

While a Treasury bill lasts less than one year from the time it is issued, a Treasury bond lasts more than one year

Bonds with a maturity between one and ten years are sometimes referred to as Treasury notes

To keep the terminology simple, many refer to all coupon-bearing Treasury instruments as Treasury bonds

U.S. Treasury bonds are defined by: The face value (or principal amount or par value), The coupon rate and the maturity date

 

The law of one price

The law of one price states that if two portfolios provide the same future cash flows, they should sell for the same price.

If the law of one price did not hold, there would be theoretical arbitrage opportunities

 

STRIPS

STRIPS is an acronym for Separate Trading of Registered Interest and Principal of Securities

STRIPS are created by investment dealers when a coupon-bearing bond is delivered to the Treasury and exchanged for its principal and coupon components

Day-count conventions describe the way in which interest is earned through time

 

Interest Rates

Investors and traders prefer to express the time value of money in terms of interest rates rather than discount factors

To understand an interest rate: We need to understand the compounding frequency used to measure it and we also need to understand the different types of interest rates

To fully describe an interest rate, we need to specify the compounding frequency with which it is measured

The compounding frequency used for an interest rate is often the same as the frequency of payments, but this is not always the case

The spot rate is the interest rate earned when cash is received at just one future time

The spot rate is also referred to as the zero-coupon interest rate, or just the “zero”

Forward rates are the future spot rates implied by today’s spot rates.

 

Some Key properties of rates

If the term structure is flat (with all spot rates the same), all par rates and all forward rates equal the spot rate

If the term structure is upward-sloping, the par rate for a certain maturity is below the spot rate for that maturity

If the term structure is downward-sloping, the par rate for a certain maturity is above the spot rate for that maturity

If the term structure is upward-sloping, forward rates for a period starting at time T are greater than the spot rate for maturity T

If the term structure is downward-sloping, forward rates for a period starting at time T are less than the spot rate for maturity T

Note: The most common swap is an agreement to exchange a fixed rate for Libor

 

Bond Yields and Return Calculations

A bond’s realized return is calculated by comparing the initial investment’s value with its final value

A bond’s yield to maturity is the single discount rate, which if applied to all the bond’s cash flows, would make the cash flows’ present value equal to the bond’s market price

 

Properties of the yield-to-maturity

When the yield to maturity is equal to the coupon rate, the bond sells for its face value

When the yield to maturity is less than the coupon rate, the bond sells for more than its face value. If time passes with no change to the yield, the price of the bond declines

When the yield to maturity is greater than the coupon rate, the bond sells for less than its face value. If time passes with no change to the yield to maturity, the price of the bond increases

If the term structure is flat with all rates equal to R, the yield to maturity is equal to R for all maturities

 

Carry roll-down

The profit or loss from a trading strategy can be decomposed into the carry roll-down, the amount resulting from interest rate changes, and the amount resulting from spread changes

The carry roll-down is usually defined as the impact of forward rates being realized (i.e., future forward rates being equal to today’s forward rates)

However, it can also be defined as the impact of the term structure remaining unchanged, or the impact of bond yields remaining unchanged

Spread changes arise from a bond’s market price moving closer to (or further away from) its theoretical price

 

Applying Duration and Convexity

One-factor risk metrics are based on the assumption that interest rate term structure movements are driven by a single factor. Examples include: DV01, duration, and convexity

Yield-based measures consider what happens to a bond price when there is a small change to its yield

Effective duration and effective convexity consider what happens when all spot rates change by the same amount. The latter is referred to as a parallel shift in the term structure

One-factor term structure shifts do not need to be parallel

The term structure shape can also change completely in a one-factor model

 

DV01

DV01 describes the impact of a one-basis-point change in interest rates on the value of a portfolio

DV01 can be calculated for any position whose value depends on interest rates

DV01 is the decrease (increase) in the price of a bond (or other instrument) arising from a one-basis-point increase (decrease) in rates

In the case of bonds, we can define DV01, duration, and convexity in terms of small changes in yields rather than small changes in all rates

This leads to some analytic results and explains the name “duration”

The DV01 of a portfolio is the sum of the DV01s of the instruments in the portfolio

 

Duration

The duration (convexity) of a portfolio is the average of the durations (convexities) of the instruments in the portfolio weighted by the value of each instrument

 

Callable bonds

A callable bond is a bond where the issuing company has the right to buy back the bond at a pre-determined price at certain times in the future

 

Puttable bond

A puttable bond is a bond where the holder has the right to demand early repayment

A puttable bond should be treated like a callable bond when calculating effective duration. In this case, the probability of the put option being exercised increases as interest rates increase

 

Modeling Non-Parallel Term Structure Shifts and Hedging

A statistical technique known as principal components analysis can be used to understand term structure movements in historical data

This technique looks at the daily movements in rates of various maturities and identifies certain factors

DV01 can be defined as the impact of a one-basis-point shift in all spot rates on the value of a portfolio

In practice, perfectly parallel shifts are rare

 

Principal components analysis

Principal components analysis shows the term structure changes observed in practice consist of the following:

A component where all rates move in the same direction, but not by exactly the same amount

A component where the term structure steepens or flattens

A component where there is a bowing of the term structure

 

Portfolio sensitivity

The sensitivity of a portfolio to shifts in the term structure can be used to calculate the standard deviation of the daily change in the portfolio value

And can therefore provide estimates of risk measures such as VaR and expected shortfall

 

Binomial Trees

Binomial trees is a valuation method widely used for pricing American-style options and other derivatives

Derivatives are valued using what is called a no-arbitrage argument

This means prices are calculated on the assumption that there are no arbitrage opportunities for market participants

The law of one price states that if portfolios X and Y provide the same cash flows at the same times in the future, they should sell for the same price

Binomial trees are a convenient way of illustrating how no-arbitrage arguments apply to derivatives

We also use binomial trees to introduce risk-neutral valuation

Risk-neutral valuation is the most important principle in derivatives pricing

 

Risk-neutrality

A risk-neutral world is one where investors do not adjust their required expected returns for risk, so that the expected return on all assets is the risk-free rate

A risk-neutral world is one where all tradable assets have an expected return equal to the risk-free interest rate

The risk-neutral valuation principle states that if we assume we are in a risk-neutral world, we get the fair price for a derivative

 

Derivatives and delta

Delta is the sensitivity of a derivative’s value to the price of its underlying stock

Trees can be constructed for valuing derivatives dependent on a non-dividend paying stock

Trees can also be constructed for valuing derivatives dependent on stock indices, currencies, and futures

When stock price movements are governed by a multi-step tree, we can treat each binomial step separately and roll back through the tree to value a derivative

For American options, it is necessary to test for early exercise at each node of the tree

 

The Black-Scholes-Merton Model

Black and Scholes used the capital asset pricing model (CAPM) to derive the relationship between the return from a stock and the return from an option on the stock

Merton used no-arbitrage arguments like those used in connection with binomial trees:

The two papers derived the same option pricing formula

The pricing formula applies to European options on non-dividend paying stocks

It can be extended to European options on stocks paying discrete dividends and to European options on other assets (such as stock indices, currencies, and futures)

It does not apply to American options, which must be valued using the binomial tree methodology

The Black-Scholes-Merton model assumes that the return from a non-dividend paying stock over a short period of time is normally distributed

 

Black-Scholes-Merton assumptions

The assumptions necessary to derive the Black-Scholes-Merton options pricing model are as follows:

  • There are no transaction costs or taxes
  • All securities are perfectly divisible
  • There are no dividends on the stock during the life of the option
  • There are no riskless arbitrage opportunities
  • Security trading is continuous
  • Investors can borrow or lend at the same risk-free rate
  • The options being considered cannot be exercised early

 

Normal and lognormal distribution

When the return on a stock over a short period is normally distributed, the stock price at the end of a relatively long period has a lognormal distribution

This means the logarithm of the stock price (and not the stock price itself) is normally distributed

A normal distribution is symmetrical and the variable can take any value from negative infinity to infinity

A lognormal distribution is skewed and the variable can take any positive value

 

Volatility

Volatility is a measure of our uncertainty about the returns provided by an investment

The implied volatility of an option is the volatility that gives the market price of the option when it is substituted into the Black-Scholes-Merton formula

There is no analytic formula for implied volatility

It must be found using an iterative trial and error procedure

 

Warrants

Warrants are options issued by a company on its own stock

If warrants are exercised, the company issues more shares, and the warrant holder buys the shares from the company at the strike price

 

Option Sensitivity Measures

The Greek letters measure different aspects of risk in derivatives portfolios:

Delta measures the sensitivity of a portfolio’s value to changes in the price of the underlying asset

Vega is the Greek letter that measures the trader’s exposure to volatility

The gamma of a stock price–dependent derivative measures the sensitivity of its delta to the stock price

The theta of an option is the rate of change in its value over time

The rho of an option measures its sensitivity to interest rates

 

Summary

Whenever you are ready, try the following links for more information:

 

Success is near,

The QuestionBank Family