Portfolio Risk: Analytical Methods

Welcome to Portfolio Risk: Analytical Methods!

Hello there! Welcome to one of the most practical chapters in the FRM Part II curriculum. So far, you have likely learned how to calculate the risk of a single asset. But in the real world, managers don't just hold one stock; they hold portfolios. This chapter is all about the tools we use to look "under the hood" of a portfolio to see which parts are driving the risk and how much risk each part contributes. Don't worry if the math looks intimidating at first—we will break it down step-by-step using simple analogies!

1. Linear vs. Non-Linear Risks

Before we dive into the formulas, we need to understand what kind of "beast" we are dealing with. Portfolios usually contain two types of instruments:

• Linear Derivatives: These are simple. If the price of the underlying asset goes up by $1, the value of the derivative moves by a fixed amount. Think of Forwards, Futures, and Swaps.
• Non-Linear Derivatives: These are "curvy." Their value doesn't move in a straight line. The classic example is an Option (because of Gamma).

Why does this matter? Because the analytical methods we use (like the Delta-Normal VaR) work great for linear risks but can be very "wrong" for non-linear risks if the price moves are large. It’s like trying to use a straight ruler to measure the length of a circle—it only works if you look at a very, very small piece of the circle.

2. VaR Mapping: Simplifying the Complex

Imagine you have a portfolio with 1,000 different corporate bonds. Calculating the risk for each one individually would be a nightmare. VaR Mapping is the process of replacing those 1,000 bonds with a few "risk factors" (like interest rates or credit spreads).

Analogy: Think of a fruit smoothie. Instead of tracking every single blueberry and strawberry, you just track the total weight of berries and the amount of sugar. You "map" the complex ingredients to a few main categories.

How Mapping Works:

1. Identify Risk Factors: Determine what moves the portfolio (e.g., equity indices, interest rate curves).
2. Calculate Sensitivities (Delta): See how much your portfolio value changes when a factor moves.
3. Map the Positions: Express your portfolio as a set of "positions" in those risk factors.

Quick Review: Mapping saves time and reduces the amount of data needed, but you might lose some specific details (called idiosyncratic risk) in the process.

3. Marginal VaR (MVaR): The "Small Change" Tool

Marginal VaR tells us how much the total Portfolio VaR changes if we add one more dollar of exposure to a specific asset.

The Formula:
\( MVaR_i = \frac{\partial VaR_P}{\partial w_i} \)
Or, a more student-friendly version:
\( MVaR_i = \frac{VaR_P}{P} \times \beta_i \)
Where \( \beta_i \) is the "beta" of the asset relative to the portfolio.

Key Concept: MVaR is a derivative (a rate of change). It tells you the impact of a tiny change in a position.

Real-World Example: Imagine you are adding salt to a soup. The "Marginal Saltiness" is how much saltier the whole pot gets if you add just one more grain of salt.

4. Incremental VaR (IVaR): The "Before and After" Tool

While Marginal VaR is for tiny changes, Incremental VaR is for big changes—like adding a whole new stock or selling an entire division.

How to calculate it:
1. Calculate the VaR of the current portfolio (\( VaR_{old} \)).
2. Add/Remove the asset and calculate the new VaR (\( VaR_{new} \)).
3. \( IVaR = VaR_{new} - VaR_{old} \)

Common Mistake: Students often confuse Marginal and Incremental. Just remember:
- Marginal = Micro (tiny change).
- Incremental = Important change (large chunk).

5. Component VaR (CVaR): The "Risk Piece"

Component VaR tells us how much of the total VaR is actually "owned" by a specific position. It is like slicing a pizza—if you add up all the slices (Component VaRs), you get the whole pizza (Total Portfolio VaR).

The Formula:
\( CVaR_i = MVaR_i \times w_i \)
(Where \( w_i \) is the dollar amount invested in that asset).

Why it’s cool:
One of the most important properties in this chapter is that:
\( \sum CVaR_i = VaR_{Total} \)
This makes it very easy to explain to a boss where the risk is coming from!

Summary Table for the "Big Three":
- Marginal VaR: Change in total VaR per unit of new investment.
- Incremental VaR: Exact change in VaR when a position is added/deleted.
- Component VaR: The portion of total VaR attributed to a specific position.

6. Risk Budgeting: Managing the "Risk Spend"

In a normal budget, you decide how many dollars to spend on rent, food, and fun. In Risk Budgeting, a portfolio manager decides how much VaR to "spend" on different asset classes.

Optimal Risk Allocation:
A portfolio is most efficient when the Ratio of Expected Return to Marginal VaR is the same for all assets.
\( \frac{E(R_1)}{MVaR_1} = \frac{E(R_2)}{MVaR_2} \)

Analogy: If you are at a buffet, you want to get the most "deliciousness" per calorie for every item you pick. If the cake gives you more joy per calorie than the salad, you eat more cake until the "joy-per-calorie" for both is equal. That is exactly what risk budgeting does with returns and VaR!

7. Key Takeaways and Tips

Did you know? If an asset has a negative Component VaR, it means that asset is actually a hedge—it is reducing the overall risk of the portfolio!

Quick Summary:
1. Mapping simplifies portfolios by using risk factors.
2. Marginal VaR uses Beta and tells us the effect of a tiny change.
3. Incremental VaR is the "Before vs. After" for large changes.
4. Component VaR slices the total VaR into pieces that add up perfectly.
5. Risk Budgeting is about equalizing the return per unit of marginal risk across all investments.

Don't worry if this seems tricky at first! The key is to remember that these are just different ways of looking at the same portfolio. Marginal is for small changes, Incremental is for big changes, and Component is for explaining the current total risk. Keep practicing the formula \( CVaR = MVaR \times Amount \), as it is a favorite on the exam!