These approaches give each observation the same weight, then no weight, in forecast
Easy to use
Problems
Can ameliorate some of these problems using an exponential weight
Weight attached to observation decays over time
for any
EWMA forecast is same as current value
But flat vol forecast not very appealing
EWMA models also take
This is implausible and not appealing
Popular alternative is a GARCH
This fits nicely with some stylised features of returns
GARCH can accommodate both of these
for
High
High
Often
GARCH
EWMA is special case with
GARCH
Run data through a preliminary filter (e.g., ARMA) to remove serial correlation
ARMA model should be as parsimonious as possible
Select particular candidate GARCH specification (e.g., based on PACs)
Estimate model using ML
Check model adequacy by testing whether standardised innovations are idd and follow assumed distribution
Pass or fail model
If model fails, try another GARCH
EWMA is special case where
Components GARCH
Factor GARCH
Correlation only good measure of dependency is returns are elliptical
Only defined if vols exist – need to check for this
Estimate correlation
where
Implied volatility
Quote ISD vs premium
These are volatilities generated from option prices
Given that other variables (price, etc.) are observable, can infer implied vol from option price (e.g., Black-Scholes)
Implied vol is a forward-looking estimator
Implied vols generally regarded as better than historical ones
Implied vols are dependent on option-pricing model
Also, implied vols only exist for assets on which options have been written
If B-S were correct, the ISDs should be constant across strike prices and maturities
If we plot ISDs agaist strike prices and maturities we obtain the implied volatility surface
Sticky strike
Sticky moneyness
The valuation formula (Margrabe Model)
Rainbow Option: exposed to two or more sources of uncertainty
Exchange rates are expressed relative to a base currency (usually USD)
The cross rate is the exchange rate between two currencies other than the reference currency
Example: let
the volatility of the cross rate is
the average correlation
all else equal, an increasing correlation in creases the total portfolio risk
A dispersion trade takes a short position in index volatility, which is offset by long position in the volatility of the index components
Estimated cov or corr matrices must be positive definite or positive semi-definite
Historical covariance matrices
Straightforward to estimate:
Drawbacks
Multivariate EWMA
Multivariate GARCH
This can be difficult
Can use methods such as orthogonal GARCH to get around some of these problems
One way to ensure PD or PSD matrices is to adjust eigenvalues, and then recover matrix from adjusted eigenvalues
Obtain eigenvalues, adjust any –ve (and maybe 0) ones using some rule
Adjusted matrix satisfies our requirements
Even if true matrix is PD (or PSD), estimated matrix might not be
Risk factors might be highly correlated
This can produce 0 or –ve estimated eigenvalues
These problems can be aggravated if covariance matrix is used for trading or risk management
Possible answers:
Holding a call option is equivalent to holding a fraction of underlying asset
Dynamic replication of a put
Dynamic replication of a long option is bound to loss money
Selling an option and dynamically hedging it using the underlying instrument
More general implications
Life of bond
When dealing with pool of mortgages, we use
Factors which affect mortgage refinancing patterns / prepayment speed
By converntion, prepayment patterns are expressed as a percentage of the PSA speed.
Project cash flows based on the prepayment speed pattern.
Prepayment risk
Dealing with the changing cash-flow pattern
The option component
Option-Adjusted Spread (OAS)
Procedures of securitization (off-balance-sheet)
Advantage of this structure
All sorts of assets (collateral) can be included in ABSs
Structure of securitization
On-balance-sheet securitizations (covered bonds or Pfandbriede in Germany)
The moral hazard problem
The adverse selection problem
cerdit rating fails for securitization that with complex structures
Securitization pushes housing prices away from their fundamental values
When the securitization markets froze, many banks and loan originators were stuck with loans that were warehoused, or held in a pipeline that was supposed to be temporary
Cash flows are redistributed to fit investors' need
This structure applies to collateralized mortgage obligations (CMOs), collateralized bond obligations (CBOs) collateralized loan obligations (CLOs), collateralized debt obligations (CDOs)
Sequential-pay tranches: defined by prioritizing the payment of principal into different tranches
Planned amortization class (PAC)
IO/PO structure: strips the MBS into two components