engle slides

7/29/2019 Engle Slides

1/68

Robert Engle

Stern School of BusinessSIEPR February 26,2009

WHAT I S H A PPEN ING TO

FIN ANC IA L M ARK ET

VOLAT I L I T Y AN D WHY ?

WHA T I S H A PPEN ING TO

FIN ANC IA L M ARK ET

VOLAT I L I T Y AN D WHY ?


2/68

RISK

A Risk is a bad future event that mighthappen.

Some risks can be avoided completely. But some risks are worth taking because

the possible benefit exceeds the possiblecosts.

Finance investigates which risks are worthtaking.


3/68

NOBEL A N SWERS

Markowitz (1952) and Sharpe(1964) andTobin (1958) received Nobel awards in

1990 and 1981 for associating risk withthe variance of financial returns.

Capital Asset Pricing Model or CAPManswer: Only variances that could not bediversified would be rewarded.


4/68

B LA CK -SCHOLES AN D M ERTON

Options can be used as insurancepolicies. For a fee we can eliminatefinancial risk for a period.

What is the right fee? Black and Scholes(1972) and

Merton(1973) developed an option pricingformula from a dynamic hedgingargument. Their answer also satisfies the

CAPM. They received the Nobel prize in 1997


5/68

I M PLEM EN T ING TH ESEMODELS

Practitioners required estimates ofvariances and covariances or equivalently

volatilities and correlations.


6/68

EST IM A T ES D IFFER FORD IFFERENT T IM E PERIODS

Volatility is apparently varying over time

What is the volatility now? What is it likely to be in the future?

How can we forecast something we never

observe?


7/68

ARCH MODEL

The ARCH model predicts the variance ofreturns on the next day.

It relies on two features of returns Volatility Clustering

Mean Reversion of Volatility Econometric Methods fit this model to

data


8/68


9/68


10/68


11/68


12/68

P lu s a n d M in u s t h r e e S i gm a

- . 1 00

-. 0 75

-. 0 50

-. 0 25

. 0 00

. 0 25

. 0 50

. 0 75

. 1 00

9 0 9 2 9 4 9 6 9 8 0 0 0 2 0 4 0 6

3 *S P V O L S P R E T -3 *S P V O L


13/68

OBSERVAT IONS

CONFIDENCE INTERVAL IS CHANGING GREEN CURVE IS APPROXIMATELY VAR

.6% RETURNS EXCEED INTERVAL LARGEST IS -6.8 SIGMA! (oct 27 1997)

MORE EXTREMES THAN EXPECTED FOR A

NORMAL BUT NOT FOR A STUDENT-T


14/68

DOES TH I S WORK I N TU RBU LEN T T IMES?

ESTIMATE THROUGH 2004 KEEPING SAME PARAMETERS,

FORECAST TO END OF SAMPLE ONEDAY AT A TIME.

DO WE SEE MULTI-SIGMA MOVES?


15/68

P lu s a n d M i n u s 3 x s i gm a u s i n g 2 0 0 4 m o d e l

- . 1 5

- . 1 0

- . 0 5

.0 0

.0 5

.1 0

.1 5

2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8

3 *D J S D 0 4 D J R E T -3 *D J S D 0 4


16/68

STANDARDIZED RETU RNS SIN CE 20 04U SING 20 0 4 EST IM A TED MODEL

0

40

80

120

160

200

240

-8 -6 -4 -2 0 2

Series: DJ RET/DJ SD04Sample 1/03/2005 10/20/2008Observations 956

Mean -0.001066Median 0.063605Maximum 3.004820Minimum -7.536694Std. Dev. 1.053278

Skewness -0.665960Kurtosis 5.993653

J arque-Bera 427.6494Probability 0.000000


17/68

WHAT WAS -7 S IGMA EV EN T ?

-8

-6

-4

-2

0

2

4

2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8

D J R E T /D J S D 0 4


18/68

SURPR IS IN G SUCCESS

Although the original application of ARCHwas macroeconomic, the big success was

for financial data.

Why does it work?

What makes volatility high?


19/68

WHY DO PR ICES CH ANGE?


20/68

B ET T ER AN SWER

Economic news on future values and risksmoves prices

Volatility is the natural response of afinancial market to new information.

News arrives in clusters. High volatility means a cluster of important

news!


21/68

VOLAT I L I T Y

Through February 20,2009

VLAB http://vlab.stern.nyu.edu


22/68

S& P 5 0 0 GARCH


23/68

ONE Y EAR TA RCH a n d V IX


24/68

ADDING Y ESTERDAY


25/68

RANGE B A SED GARCH U SINGA SYMM ETR IC M EM FOR DA X


26/68

CORPORA TE BONDS


27/68

MSC I WORLD INDEX


28/68

MSCI EM ERGING M ARK ET INDEX


29/68

COMMOD IT I ES


30/68

ENERGY , F IN A NCE, T ECHNOLOGY


31/68

SECTOR CORRELA T ION S


32/68

I N T ERN A T IONA L CORRELA T IONS


33/68

WHERE I S VOLA T I L I T Y TODA Y ?

For most assets, volatility last fall wasdramatically above levels since 1990 but

is now somewhat lower. In the US, I think this is due

A) Macroeconomic uncertainty

B) Credit problems particularly associatedwith securitized debt.


34/68


35/68

TH E SPL IN E GARCH MODEL OFLOW FREQUENCY V OLA T I L I TY A ND IT S M ACROECONOM IC

CAUSES

Robert Engle and Jose Gonzalo RangelReview of Financial Studies 2008


36/68

MODEL LOW FREQUENCYVOLAT I L I T Y

For what countries is this greatest? For what time periods is it greatest?

What macroeconomic variables areassociated with volatility?


37/68

WHA T M AK ES F IN AN CIA L M ARK ET VOLAT I L IT Y H IGH ?

High Inflation Slow output growth and recession

High volatility of short term interest rates High volatility of output growth

High volatility of inflation

Small or undeveloped financial markets

Large countries


38/68

WERE WE PREPARED?


39/68

TH REE VOLAT I L I T Y EP ISODES


40/68

DOW J ONES 1 9 2 8 -2 0 0 8

-.3

-.2

-.1

.0

.1

.2

30 40 50 60 70 80 90 00

D J R E T

DJ CLOSE


41/68

400

200

160

120

100

80

60

40

20

1930 1932 1934 1936 1938 1940

DJ CLOSE

VOLCOMP


42/68

0

20

40

60

80

100

120

1930 1932 1934 1936 1938 1940

VOLCOMP


43/68

DJ VOLAT I L I T Y 1 9 8 0 -1 9 9 0

0

20

40

60

80

100

120

140

160

80 81 82 83 84 85 86 87 88 89 90

D J V O L _ C O M P


44/68

AND FOR 1 9 9 8 -2 0 0 8 ? WHA T CAN WE EX PECT ?


45/68


46/68


47/68

DJ VOLAT I L I T Y 1 9 9 8 -2 0 0 8

0

1 0

2 0

3 0

4 0

5 0

6 0

7 0

9 8 9 9 0 0 0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8

V O L C O M P P E R M


48/68

TH E R I SK OF WAR an d TERROR ISM


49/68

A LONG RUN R I SK

Deteriorating Global Economy Increasing income differential between

rich and poor countries

Rising fundamentalism

Rising social unrest

Increase the risk of War and Terrorism


50/68

DEPRESSED A SSET PR ICES

Long run risks lower asset prices asinvestors are more cautious.

This raises the cost of doing business andraising capital

This reduces income of entrepreneurs

And costs jobs


51/68

WHAT TO DO?

PROMOTE PEACE MANY, MANY APPROACHES THROUGH POLITICS,

SCIENCE, MEDICINE, CULTURE, EDUCATION, LAW

SOME ECONOMIC PROPOSALS: TRADE CAPITAL FLOWS

BUILD ECONOMIC INTERDEPENDENCES

FIGHT POVERTY REFORM EDUCATION to show value in cooperation

PEACE PERMITS PROSPERITY


52/68

BENEF I TS

Reducing future risk of war Yields benefits today by

Improving business and stock marketvaluations and

Creating jobs


53/68

V ERY LONG RUN R ISK S ! ARE WE READY FOR TH ESE?


54/68

GLOBA L OVERHEA T ING WHA T A RE TH E R I SK S?

Scientific evidence seems clear that theclimate is changing.

CO2

concentrations are rising rapidly

Glaciers and polar ice are melting

Warmest years on record are almost all within

10 years.

But what are the costs? Scientific

evidence is not precise.


55/68

ECONOM IC COSTS

THE GLOBAL ECONOMY WILL BE UNABLETO PRODUCE AS MUCH IN THE FUTURE ASIT WOULD WITHOUT CLIMATE CHANGE

TAXES WILL BE RAISED TO PAY FORPUBLIC EFFORTS TO MITIGATE THESECOSTS

COMPANIES WILL HAVE EXTRA COSTS OFDOING BUSINESS SO PROFITS WILL BE

LOWER.


56/68

FI NAN CI A L M ARK ET EVA LU AT ION OF CL IMA TE R ISK . I S CL IM A TE R ISK PRICED?

Can we see evidence of climate risk in financialmarkets? We would expect that stock prices would be

depressed by climate risk. If it is a risk for all stocks, then it would imply

simply a lower price. If it is more of a risk for some companies or

countries than others, then they would havebigger discounts.

If these risks are not priced, then there could be

profitable portfolio strategies.


57/68

A SOLU T ION


58/68


59/68


60/68

A SOLU T ION

Most Economists believe the best solutionto global overheating is a comprehensivetax on carbon emissions and othergreenhouse gases.

Only if it is comprehensive will it encourage

alternative energy solutions Only if it is comprehensive will efforts to avoid

the tax be socially beneficial.


61/68

TA X REVENUE MU ST B E PARTOF TH E AN A LY SI S

Send a check to every resident for an equalfraction of the total revenue. This compensates everyone equally so it offsets the hardship

of an emissions tax for low incomes.

Send the checks first. It will enable people to buy fuel efficient cars, insulate homes,

improve appliance efficiency. This will stimulate industries thathave been badly hurt by the financial crisis.

After recession is over, invest passively in a sovereignfund to support social security and public health care.


62/68

H ig h O i l Pr i c e s a r e a Go o d T h i n g !

These encouraged consumers andindustry to use less oil

Driving in the US was down

Hybrid Cars were selling and SUVs were not

House prices in the suburbs were declining

more than in the central city Ridership on public transportation was up


63/68

B u t t h i s i s N o t En o u g h .

Oil prices are falling again. Coal is still a cheap and dirty alternative.

Entrepreneurs with ideas for alternativeenergy sources cannot be confident thatenergy prices will stay high.

Businesses and Consumers will hesitatebefore making energy investments


64/68


65/68

CONCLUS ION

Make sure you take only the risks youintend to take

Keep an eye on long run risks

Policy makers remember: reducing longrun risks gives benefits today


66/68


67/68

The first European conference will be held in GenevaJune 10th, 11th, and 12th, 2009.

Information about submitting papers can be found at the following webpage:

http://www.nyu.edu/sofie/

FoundingCouncil

John CampbellFrank DieboldRobert Engle

Ronald GallantRen Garcia

John GewekeEric Ghysels

Christian GourirouxClive Granger

Lars Peter HansenWolfgang HrdleRavi Jagannathan

Eric RenaultGeorge Tauchen

Halbert White

The Society is a global network of academics and practitionersdedicated to the fast-growing field of financial econometrics.

The Society will be associated with theJournal of Financial

Econometrics.

The Society for Financial EconometricsThe Society for Financial Econometrics


68/68

engle slides

Documents