![Page 1: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/1.jpg)
Class 9: Time-Varying VolatilityFinancial Markets, Spring 2020, SAIF
Jun Pan
Shanghai Advanced Institute of Finance (SAIF)Shanghai Jiao Tong University
March 26, 2020
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 1 / 27
![Page 2: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/2.jpg)
Outline
The importance of measuring market volatility:▶ Portfolio managers performing optimal asset allocation.▶ Risk managers assessing portfolio risk (e.g., Value-at-Risk).▶ Derivatives investors trading non-linear contracts.
Volatility σ can be better measured than expected returns µ:▶ Backward looking: estimate volatility using historical data.▶ Forward looking: from derivatives prices.
Estimating volatility using financial time series:▶ SMA: simple moving average model (traditional approach).▶ EWMA: exponentially weighted moving average model (RiskMetrics).▶ ARCH and GARCH models (Nobel Prize).
EWMA for covariances and correlations.
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 2 / 27
![Page 3: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/3.jpg)
Modern Finance
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 3 / 27
![Page 4: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/4.jpg)
The Evolution of an Investment Bank
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 4 / 27
![Page 5: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/5.jpg)
Derivatives Losses by Non-Financial Corporations in 1990s
Orange County: $1.7 billion, leverage (reverse repos) and structured notes
Showa Shell Sekiyu: $1.6 billion, currency derivatives
Metallgesellschaft: $1.3 billion, oil futures
Barings: $1 billion, equity and interest rate futures
Codelco: $200 million, metal derivatives
Proctor & Gamble: $157 million, leveraged currency swaps
Air Products & Chemicals: $113 million, leveraged interest rate and currency swaps
Dell Computer: $35 million, leveraged interest rate swaps
Louisiana State Retirees: $25 million, IOs/POs
Arco Employees Savings: $22 million, money market derivatives
Gibson Greetings: $20 million, leveraged interest rate swaps
Mead: $12 million, leveraged interest rate swaps
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 5 / 27
![Page 6: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/6.jpg)
Measuring Market Risk
By early 1990s, the increasing activity in securitization and the increasingcomplexity in the financial instruments made the trading books of many investmentbanks too complex and diverse for the chief executives to understand the overall riskof their firms.
Market risk management tools such as Value-at-Risk are ways to aggregate thefirm-wide risk to a set of numbers that can be easily communicated to the chiefexecutives. By the mid-1990s, most Wall Street firms have developed riskmeasurement into a firm-wide system.
Daily estimates of market volatility, along with correlations across financial assets,constitute the key inputs to Value-at-Risk. JP Morgan’s RiskMetrics usesexponentially weighted moving average (EWMA) model to estimate the volatilitiesand correlations of over 480 financial time series in order to construct avariance-covariance matrix of 480x480.
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 6 / 27
![Page 7: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/7.jpg)
Daily Returns on the S&P 500 Index
1970 1980 1990 2000 2010 2020-25
-20
-15
-10
-5
0
5
10
15Daily SPX Returns (%)
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 7 / 27
![Page 8: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/8.jpg)
The Simple Moving Average Model
Unlike expected returns, volatility can be measured with better precision usinghigher frequency data. So let’s use daily data.
Some have gone into higher frequency by using intra-day data. But micro-structurenoises such as bid/ask bounce start to dominate in the intra-day domain. So let’snot go there in this class.
Suppose in month t, there are N trading days, with Rn denoting n-th day return.The simple moving average (SMA) model:
σ =
√√√√ 1
N
N∑n=1
(Rn)2
To get an annualized number: σ ×√252. (252 trading days per year).
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 8 / 27
![Page 9: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/9.jpg)
The Monthly SMA Volatility Estimates
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 9 / 27
![Page 10: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/10.jpg)
The Precision of the SMA Estimates
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 10 / 27
![Page 11: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/11.jpg)
Time-Varying Volatility and Business Cycles
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 11 / 27
![Page 12: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/12.jpg)
SMA Volatility and Option-Implied Volatility
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 12 / 27
![Page 13: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/13.jpg)
Review: SMA Volatility Estimates
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 13 / 27
![Page 14: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/14.jpg)
Exponentially Weighted Moving Average Model
The simple moving average (SMA) model fixes a time window and applies equalweight to all observations within the window.
In the exponentially weighted moving average (EWMA) model, the more recentobservation carries a higher weight in the volatility estimate.
The relative weight is controlled by a decay factor λ.
Suppose Rt is today’s realized return, Rt−1 is yesterday’s, and Rt−n is the dailyreturn realized n days ago. Volatility estimate σ:
Equally Weighted Exponentially Weighted√√√√ 1
N
N−1∑n=0
(Rt−n)2
√√√√(1− λ)N−1∑n=0
λn (Rt−n)2
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 14 / 27
![Page 15: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/15.jpg)
EWMA Weighting Scheme
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 15 / 27
![Page 16: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/16.jpg)
SMA and EWMA Estimates after a Crash
Source: J.P.Morgan/Reuters RiskMetrics — Technical Document, 1996
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 16 / 27
![Page 17: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/17.jpg)
Computing EWMA recursively
One attractive feature of the exponentially weighted estimator is that it can becomputed recursively.
Let σt be the EWMA volatility estimator using all the information available on dayt − 1 for the purpose of forecasting the volatility on day t.
Moving one day forward, it’s now day t. After the day is over, we observe therealized return Rt .
We now need to update our EWMA volatility estimator σt+1 using the newly arrivedinformation (i.e. Rt). It turns out that we can do so by
σ2t+1 = λσ2
t + (1− λ)R2t
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 17 / 27
![Page 18: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/18.jpg)
Decay factor, Strong or Weak?
A strong decay factor (that is, small λ) underweights the far away events morestrongly, making the effective sample size smaller.
A strong decay factor improves on the timeliness of the volatility estimate, but thatestimate could be noisy and suffers in precision.
On the other hand, a weak decay factor improves on the smoothness and precision,but that estimate could be sluggish and slow in response to changing marketconditions.
So there is a tradeoff.
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 18 / 27
![Page 19: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/19.jpg)
Fast, Medium, and Slow Decay
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 19 / 27
![Page 20: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/20.jpg)
ARCH and GARCH models
The ARCH model, autoregressive conditional heteroskedasticity, was proposed byProfessor Robert Engle in 1982. The GARCH model is a generalized version ofARCH.
ARCH and GARCH are statistical models that capture the time-varying volatility:
σ2t+1 = a0 + a1 R
2t + a2 σ
2t
As you can see, it is very similar to the EWMA model. In fact, if we set a0 = 0,a2 = λ, and a1 = 1− λ, we are doing the EWMA model.
This model has three parameters while the EWMA has only one. So it offers moreflexibility (e.g., allows for mean reversion and better captures volatility clustering).
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 20 / 27
![Page 21: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/21.jpg)
The Nobel-Prize Winning Model
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 21 / 27
![Page 22: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/22.jpg)
Review: EWMA Volatility Estimates
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 22 / 27
![Page 23: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/23.jpg)
EWMA Covariances and Correlations
Our goal is to create the variance-covariance matrix for the key risk factorsinfluencing our portfolio.
For the moment, let’s suppose that there are only two risk factors affecting ourportfolio.
Let RAt and RB
t be the day-t realized returns of these two risk factors. Thecovariance between A and B:
covt+1 = λ covt + (1− λ)RAt × RB
t
And their correlation:corrt+1 =
covt+1
σAt+1σ
Bt+1
,
where σAt+1 and σB
t+1 are the EWMA volatility estimates.
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 23 / 27
![Page 24: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/24.jpg)
Negative Correlation between RM and ∆VIX
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 24 / 27
![Page 25: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/25.jpg)
Correlation between SPX and SSE
1995 2000 2005 2010 2015 2020-20
-10
0
10
20
30
40
corr
elat
ion
(%)
EWMA Correlation of SPX and SSE using 5-Day Returns
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 25 / 27
![Page 26: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/26.jpg)
From Volatility Estimates to VaR
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 26 / 27
![Page 27: Class 9: Time-Varying Volatilityen.saif.sjtu.edu.cn/junpan/FMar_2020/slides_Volatility.pdf · Outline The importance of measuring market volatility: Portfolio managers performing](https://reader036.vdocuments.us/reader036/viewer/2022062603/5f02b1007e708231d405874b/html5/thumbnails/27.jpg)
The Main Takeaways
Financial Markets, Spring 2020, SAIF Class 9: Time-Varying Volatility Jun Pan 27 / 27