Beruflich Dokumente
Kultur Dokumente
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Institute of Mathematical Statistics is collaborating with JSTOR to digitize, preserve and extend access to
Statistical Science.
http://www.jstor.org
Statistical Science
1992, Vol. 7, No. 1, 69-122
69
70 L. M. BERLINER
are now beginningto make contributionsalong specifyingthe initial conditionis in the sixth
these lines, the methodsdescribedin Section 3 decimalplace. This sortofbehavior,knownas sen-
have been developedprimarilyby mathematicians sitivity to initial conditions, is one of the key
and physicists.In Section4, I discusssomepossible componentsof chaos. To amplifyon this phe-
strategiesformethodsofchaoticdata analysisbased nomenon,the firstframeof Figure 2 presentsdot
on main streamtechniquesforstatisticalmodeling plots, at selected time values, of the dynamical
and inference.Finally,Section5 is devotedto gen- system correspondingto 18 initial conditions
eral remarksconcerningstatisticsand chaos. equally spaced in the interval [0.2340, 0.2357].
There are two messages in this plot. First, note
that the images of these 18 points are quickly
2. MATHEMATICS,PROBABILITY AND CHAOS attractedto the unit interval.Second,the initial
2.1 The Complexityof NonlinearDynamical
conditionsappear to get "mixed" up in an almost
Systems
noncontinuousmanner.(However,forthe logistic
map, xt is, of course,a continuousfunctionof xo
A simpledeterministic dynamicalsystemmaybe forall t). The secondframeofFigure2 is a scatter-
definedas follows.For a discretetime index set, plot of the values of the logisticmap after2000
T = {0, 1, 2, ... }, considera timeseries { x,; te T}. iterates against the correspondinginitial condi-
Assume that xo is an initial conditionand that tionsfor4000 initials equally spaced in the inter-
xt+1= f(xt), forsome functionf that maps a do- val [0.10005, 0.3]. There is clearlyessentiallyno
main D into D. (D is typicallya compactsubsetof meaningfulstatementsabout the relationshipbe-
a metricspace). Chaoticbehaviormayarise when f tween x2000and xo, even though x2000is a well-
is a nonlinearfunction. definedpolynomialfunction(of admittedlyhigh
To begin,somenumericalexamplesforone ofthe order) of xo. (Note that presentingthis graph
morepopular examples of dynamicalsystems,the for 2000 iterates is a bit of "overkill." Corre-
logisticmap, are given. The dynamicalsystemis spondingscatterplots aftereven a 100 or so iterates
obtainedbyiteratingthefunctionf(x) = ax(l - x), wouldlook quite the same.)
where a is a fixedparameterin the interval[0,4].
Let xo be an initialpointin the interval[0,1];note 2.1.1 Some mathematicsfor nonlinear dynamical
thatthenall futurevalues ofthe systemalso lie in systems
[0,1].To get a bit ofthe flavorofthismap,example This discussionis intendedto providesomeflavor
computationsare presentedforan importantvalue of the mathematicsconcerningthe appearance of
of a: namely,a = 4.0. Figure 1 presentstimeseries complexor chaoticbehaviorin nonlineardynami-
plots of the first500 iterates of the logisticmap cal systems.The presentationis a bit quick, and
corresponding to the initial values 0.31, 0.310001 untilSection2.1.3, considersone-dimensionalmaps
and 0.32. The firstthing to notice about these only. More complete details maybe foundin Collet
series is that their appearance is "complex." In- and Eckmann(1980),Rasband (1990) and Devaney
deed,one mightbe temptedto suggestthese series (1989). We begin by consideringthe long-run
are "random." Also, despitethe similarityin the behavior of a dynamicalsystemgeneratedby a
initial conditions,visual inspectionof the series nonlinearfunctionf. The studybegins with the
indicatesthat theyare not quantitativelysimilar. considerationof fixedpoints of f; namely,those
To make the point,I have includedscatterplotsof pointsthat are solutionsto f(x) = x. The key re-
these series,matchedby time.The first25 iterates sult in this contextis the followingproposition.
ofthe maps in theseplotsare indicatedby a differ- Using conventionalnotation,let ffn(.) denotethe
ent symbol fromthe rest. Points falling on the 450 n-foldcomposition of f.
line in theseplotssuggesttimevalues at whichthe
corresponding values ofthe systemsare quite close. PROPOSITION 2.1. Let p be a fixed point of f. If
We see that quite early in time,the three series I f'(p) I < 1, then thereexistsan open intervalU
"predict"each otherreasonablywell. However,the about p such that, forall x in U, limn-+o fn(x) = p.
similarityin the series diminishesrapidlyas time
increases.(The rate ofthis "separation" is in fact Under the conditionsof this proposition,p is an
exponentialin time.) Note' that, except for very attractingfixed point and the set U is a stable set.
early times, the series correspondingto xo = i i
It is also truethat if f'(p) > 1, p is a repelling
0.310001 is really no betterat predictingthe 0.31 fixedpoint. (In these two cases, p is said to be
series than is the 0.32 series. Also, predictions hyperbolic; if I f'(p) I = 1, p acts as a saddle point
based on xo = 0.31 when the correctvalue is and moredelicate analyses are required.The rest
0.310001 are verypoor,even thoughthe errorin ofthis discussionfocuseson hyperbolic points.)
CHAOS 71
0.6 o.6
A
0.21 0.
100 200 300 400 100 200 300 400
series series
1.0 -
0.6
0.4
C
series
(i)
1.0 x x1 .0 0 x x 0 X
0
x
0 0
0.8 - 0.8
0.6 - X 0.6 x
| R | 0~~~~
O~~~~04 x
A 0.4 ?
0
X A 04 t x
0.2 -- 0 x 0.2
op x Qo0 x x
, X
0191t' XX-tx 00X0
0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0
B C
0.8 .. . 0.8
0.6 . ;
'* 0.6
A ... ... A
0.2 t. 0.2 t.;
0 0 0. . . . . . .
0.2 0.4 0.6 0.8 1.0 0.2 0.4 0.6 0.8 1.0
B (ii) C
FIG.1. Examples ofthe logistic map: a = 4.0. Initial conditions:(A) x0 = .31; (B) X0 = .310001;(C) X0 = .32. (i) Time seriesplots for
times 26 - 500. (ii) Scatterplotsmatched by time. In topplots, 0 denotes times 1 - 15 and x denotes times 15 - 25; bottomplots show
times 26 - 500.
72 L. M. BERLINER
1.0 8 0
0
?+
* x
0.8 4 X > + 0~~~~~~ x
o x X +
0 ~~~x+
0.6 0+
O
0 x 0~~~~~~~~~~~~~~
+
t + 0 X
0.4- +
o
x~~~~~~ o
0 O +
0 0
+ +
0.2 +
~~0
x
x
20 30 40 50 100 200
1.0 d.~~~~~~~~~~~~
0.0~~~~~~~~~~~~~~~~~~~5 (i)
%~~~~~~~~~~~~~~%.s
0.40.2t/
8. *|;. -fi;!rX
.:; .,a.
P I~~~~~~3~~.
' .
6 * _
xo
(ii)
FIG.2. Mixingbehaviorofthelogisticmap: a = 4;0. (i) Dot plotsfor18 initialconditions.0 denotesxo = .2340,.2341, .2342,.2343,
.2344, .2345; x denotesxo = .2346, .2347, .2348, .2349, .2350, .2351; + denotesxo = .2352, .2353, .2354, .2355, .2356, .2357. (ii)
Scatterplotoflogisticmap at time2000 againstxo for4000 xo's in [.10005,.3].
ing to the threepointsofthe attractor;that is, the 3.5 3.6 3.7 3.8 3.9 4
threeattractingfixedpointsof f3. For example,my a
computationsindicate that the stable set corre- FIG. 4. Iteratesofthelogisticmap.
spondingto the point 0.1561... is approximately
I, = [0.14545,0.163571.This estimatecan be veri- tivityto initialconditionscan occurwithLebesgue
fiedby checkingthat f3(X) E11 forxE 1I. This cri-
terionwas numericallysatisfiedfor2000 equally measure0. Maps forwhichall pointsmustseparate
spacedpointsin I,. underiterationare said to be expansive.However,
To summarizethe asymptoticbehavior of the expansivenessis typicallytoo restrictiveformost
logisticmap, considerthe plot in Figure 4. This maps. (For example,considerthelogisticmap when
plot is intendedto indicate the attractorof the a = 4. The initial conditions xo and 1 - xo result
in identicalrealizationsofthemap.Therefore, since
logisticmap as a functionof a. For a gridofvalues
of a from3.45 to 4, iterates 101 to 221 of the xo can be chosenarbitrarily closeto 0.5, expansive-
logistichave been plotted.The result,knownas an ness cannotbe claimed).However,the set ofall x
orbitdiagram,is an interesting, complexobject. leading to any periodicbehaviorwhen a = 4 is a
set of Lebesgue measure zero. This sort of phe-
chaos?
2.1.2 Whatis mathematical nomenonis related to anothercomponentof the
mathematicaldefinition ofchaos,namelythe set of
There does not appear to be a universallyac- x's leadingto periodicbehavioris densein D. That
cepted,mathematicaldefinition ofchaos. Thereare is, complex,aperiodicbehaviorcan arise despite
differentwaysto quantifywhatone mightmean by the existenceof denselydistributedopportunities
complexor unpredictablebehavior. The primary forwell-ordered behavior.This is implicitin the
conceptappears to b& the notionsensitivityto ini- claim of Li and Yorke (1975) that "period three
typicallyquantifiedas:
tial conditions, implieschaos." I thinkthe way mostpeoplewould
like to interpretsensitivityis as if the map were
DEFINITION2.1. f: D -+ D. displays sensitivityto almost everywhere,typically,in the sense of
if
initialconditions there exists 6 > 0 such that for Lebesguemeasure,expansive.
any x in D and any neighborhoodV of x, there
exists a y in V and n 2 0 such that fn(x)- D is almost everywhere
DEFINITION 2.2. f:D -
fn(y)I > 6 expansiveifthereexists 6 > 0 suchthatforalmost
all x in D and almost all y in D, there exists
suggeststhat thereexist pointsar-
t'his definition n 2 0 suchthat I fn(x)-fn(y) I > 6.
bitrarilycloseto x thatseparatefromx duringthe
time evolutionofthe dynamicalsystem.However,
AlthoughI have not located discussionof such a
the definitiondoes notsay all pointsmustseparate,
there may be relationshipsto some of
definition,
apparentlyleaving open the possibilitythat sensi-
the definitionsrelatingchaos and randomnessdis-
cussedbelow.
Anothermathematicalconceptassociated with
definitionsof chaos intuitivelyinvolvesthe rich-
I .. . .. . . .I..... . ... .. .... . . . ness ofchaoticpaths.
0.30
DEFINITION 2.3. f:D-+ D is topologicallytransi-
0.21 0.24 0.27
tiveif forany pair of open sets U, V in D there
0.18
I1
(Definition2.3); and (iii) periodicpointsare dense.
Other related definitionsof chaos (positive Lia-
punov exponents,as introducedin Section 3.1.1,
and the existenceof continuousergodicdistribu-
tions, as introducedin Section 2.2.3) involveno- 300 400 500
tions of ergodictheory.See Collet and Eckmann
0 100 200
(ii)
(1980) forfurtherdiscussion. FIG. offirst2000 iter-
5. PlotsforHenonmap. (i) Scatterplot
ates. (ii) First500 iteratesofx.
2.1.3 Dissipative systemsand chaos
Much ofthe complexbehaviorofthe logisticis a
resultofits noninvertibility.Indeed,noninvertibil- Carleson(1991).] To get a feelforthis map,Figure
is
ity required to observe chaos forone-dimensional 5 presentsa scatterplotofthe first2000 iterations
dynamicalsystems, as defined here. However,ev- (the attractor)ofthe map, as well as a time series
erywhere invertible maps in two or more dimen- plot of the "x" series. These computationswere
sions can also exhibit chaoticbehavior. Among the based on the conditionsa = 1.4, b = 0.3, xO= 0.4
manyinterestingfacetsof dynamical systems, one and yo = 0.3.
area that receivesmuch attention is the study of This example illustratestwoimportant aspectsof
strangeattractors.The basis issue is the long-run chaotic behavior. First,note that the complex geo-
behaviorofthe system.As time proceeds, the tra- metrical structure of the Henon attractor. This ob-
jectoriesofsystemsmaybecometrapped in certain ject is of fractal dimension. Such objects appear in
boundedregionsof the state space of the system. the study of many dynamical systems. The mathe-
As notedeven forthe logisticmap, these trapping maticsthat suggestsuch behaviorare as follows.
regionsor attractors can display remarkable oddi- The Henon map, viewedas a transformation from
ties. An importantexample in two dimensions is R2 to R2, has Jacobian equal to -b. If 0 < b < 1,
the Henon map. This map can display the property we make the geometrical observation that Henon
ofhavinga strangeattractor;that is, the attractor map contractsthe areas of sets to whichit is ap-
"appears to be locallythe productof a two-dimen- plied. More generally,such maps are said to be
sional manifoldby a Cantorset." This quote,along dissipative. (Maps that maintainarea underitera-
with a motivationof the map, may be found in tion are conservative.) Intuitively,the complex
Henon (1976). Also, see the previousreferences for limiting behaviorofchaotic,dissipativedynamical
discussion.The Henon map is givenby the follow- systems is the resultoftwo competingmathemati-
ing equations: cal trends. Dissipativenesssuggeststhat iterates
tend to collapse to sets ofLebesgue measurezero.
(2.3) xt1 = + yt - ax2 and Yt+1 = bxt However,an effectof chaos is to prohibitperiodic
behavior.The naturalresultsconsistentwiththese
forfixedvalues of a and b 'and t = 0, 1, .... This twophenomenais forthe systemto be attractedto
invertiblemap can not onlypossess strangeattrac- an inflnite,singularset ofLebesgue measurezero
tors,but also display strongsensitivityto initial (in an appropriatemanifoldofRk). Such attracting
conditions.as encounteredearlier. [The rigorous sets are knownas strange attractors. (The above
verificationofmanyofthe propertiesofthe Henon heuristicsare notcomplete.For example,conserva-
map is actually very difficult;see Benedicksand tive systemscan displaychaoticbehaviorwithout
76 L. M. BERLINER
being attracted to singular attractors. Rather, here. The reader can find a valuable discussion of
conservative systems can have attractors that are the basics in Press, Flannery, Teukolsky and Vet-
colorfully called "fat fractals"; that is, complex terling (1986).
geometrical objects that are essentially "space-fill- The followingexample of (2.5) is used in Section
ing." Such systems will not be considered further 4. The differentialequation, known as the Lorenz
.n this article. The reader may consult the refer- system, is extremely popular in the literature on
ences fordiscussion.) chaos. The system,given component-wise,is
Second, note that there is a relationship between
the roles of time and dimension in the definitionof dx/dt = a(y - x),
dynamical systems. In particular, (2.3) can be writ- (2.6) dy/dt = -xz + rx - y,
ten as a one dimensional relationship if we allow
dz/dt = xy - bz,
consideration of more lags of time:
where a, r and b are constants. Lorenz (1963)
(2.4) x = 1 + bxt1 - ax.
considered this systemas a rough approximationto
Statisticians familiar with more conventional time aspects of the dynamics of the Earth's atmosphere.
series modeling might see a kinship between (2.4) Figure 6 presents plots in phase space of a numeri-
and a "nonlinear autoregressive model of order 2." cal approximation of a solution to (2.6) where a =
I will return to such concepts in Section 4. 10, r = 28 and b = 8/3. (I tried to indicate why the
attractor is nicknamed "The Butterfly.") For this
2.1.4 Continuous time dynamical systems:Differ- famous choice of the parameters, the solutions (see
ential equations Figure 7) display sensitive dependence to initial
Continuous time dynamical systems arise natu- conditions and "unpredictable" fluctuations. For
rally in many applications in which the time almost all initial conditions,orbits are attracted to
evolution of the quantities of interest,composing a the object displayed in the various panels of Figure
k-dimensionalvector x(t), are modeled via differen- 6. Note that I have plotted points fromthe discrete
tial equations. In particular, consider a initial value time numerical approximation in this figure.
problem where x(O) is an initial condition and the "True" solutions to (2.6) are continuous and are
dynamics ofthe system are quantified by the differ- attracted to a "continuous," yet strange attractorof
ential equation Lebesgue measure zero, since the Lorenz system is
dissipative.
(2.5) dx(t)/dt = F(x(t)), t > 0.
2.2 Randomness and Chaos
The value x(t) describes the state of the system at This section reviews various relationships be-
time t; the domain of possible values of x( ) is tween chaos and randomness. The key ideas
called the phase space. A specific solution to (2.5), involve the interrelations between sensitivity to
"plotted" in the phase space, is known as an orbit. initial conditions, uncertainty modeling and er-
As in the case of discrete time, solutions to (2.5) godic theory.The discussion emphasizes ideas, but,
can be chaotic in the sense that some views of the forthe sake of brevity,not rigor.
solutions may appear "random," solutions display
sensitivity to initial conditions and, in the case 2.2.1 Uncertainty,chaos and randomness
of dissipative systems, now indicated when
The main topic ofthis section is how uncertainty,
1 /a xi < 0, orbits are attracted to strange
aFi
especially in the presence of complexity,naturally
=
attractors.
leads to the use ofrandom or probabilisticmethods.
Traditional methods for numerically solving dif-
I will begin with a historical perspective. An early
ferential equations typically involve discrete time
and persuasive suggestion that deterministicmod-
approximations. The simplest method, in the con-
els may be of limited value is the followingdiscus-
text of (2.5) uses the approximation x(tn+l) =
sion of Laplace, circa 1800, (from Laplace, 1951,
h F(x(tn)) + x(tJ) where the stepsize, h, is small
page 4):
and tn+1 = h + tn. Thus, numerical solutions to
differentialequations are themselves typically dis- Given for one instant an intelligence which
crete time dynamical systems. For the actual could comprehend all the forces by which na-
computationsin this article, I used a more sophisti- ture is animated and the respective situation of
cated approximation known as the four-point the beings who compose it-an intelligence suf-
Runge-Kutta method. This method appears to be ficiently vast to submit these data to analy-
considered a standard method for solving differen- sis-it would embrace in the same formulathe
tial equations. Further details are not relevant movements of the greatest bodies of the uni-
CHAOS 77
W: ~ ~~IZ ) j
r~~
oV l ~~~ S[ljii l*jl j-u l ji ~ ji9
10
l l l
30~~~~~~~~~~~~~r
2000 4000 6000 80000
TINE
20
. ..'-
* ..*.
* -...
... . . -.
--e. .
-314t
-it-- ~~ ~ ~ ~ ~ : *
2000 4000 6000 8000
TIME
(i) FIG. 7. Time seriesplots fortheLorenz system.
beyondthe scopeofthispaper.
To see the role ofergodictheoryin deterministic
chaos,we will need a bit offormalism.Considera 0.0000 0.1 625 0,3250 0.4875 0.6500 0.81 25 0.9750
{wr/x(1 - x) }I-. It is easy to checkthatthe logis- FIG. 8. Example of ergodic behavior: Logistic map, a = 4.0. (i)
Histogram of 4000 iteratesof xo = .20005. (ii) Histogram of the
tic function, f(x) = 4 x(1 - x), is invariant with logistic map at time 2000 for 4000 xo's in [.10005, .30005].
respectto P.
A fewmoredefinitions are neededto relatethese
ideas to dynamical systems. For an invertible show that exceptforX, forwhich P(X) = 1, any
transformation f, a subset AeF is invariantif invariantsubset must be denumerableand thus
f `(A) = A. For a noninvertible f,invarianceof A have P-measure0. Ergodicbehavioris suggested
means ft(A) = A forall t > 0. Further,f is said to by consideringsamplepaths ofthe logistic.Figure
be ergodic if for every invariant subset, AeF, 8 (i) presentsa histogramof 4000 iteratesof this
P( A) = 0 or 1. That is, if f is ergodic,samplepaths map, beginning at the initial condition xo =
of the dynamicalsystemobtainedfrom f do not 0.20005. Figure8 (ii) presentsthe histogramofthe
becometrappedin propersubsetsofthe supportof values of the logisticmap at time 2000 for4000
P, but rathermix overits support.(Note the corre- different initial conditions.In both cases, we see
spondenceto topologicaltransitivityin Definition the appearanceofsomethinglike the arc-sineden-
2.3.) Furthermore, if f is ergodicwith respectto sity.In the construction ofFigure8 (ii) I have used
two probabilitymeasures P, and P2 on the same what is perhapsthe mostimportantfeatureofthe
measurespace,theneitherP, = P2 or P1 is orthog- applicationofthe ergodictheoremto deterministic
onal to P2. With this structure,we can state a dynamicalsystems.The key point is to note the
simpleversionofthe ergodictheorem: powerofthe "almostsure" convergence ofthetheo-
rem.That is, the initial condition xo the system
of
If f is measure-preservingand ergodic on (X, F, need notbe randomlygenerated according to P for
P) and Y is any random variable such that E( I Y I) ergodicity to applyto the resulting dynamical sys-
< oo, then tem. Indeed,the initial condition need not be "ran-
domlygenerated."We must onlyavoid sets of P-
1n measure zero. (For the arc-sinedistribution, this
'(xo)) -+Ep(Y) a.s.(P), asn -oo. Furthermore, for
n i=l Y(f means Lebesgue measure zero.)
E
a
any logisticwith large (a > 3.83 ... ) enough to
To illustratethisresult,again considerthe logis- admit periodiccycles of all orders, each periodic
tic map with a = 4 and let P correspond to the arc cyclegeneratesa discreteergodicdistribution that
sin law. To see that f is ergodic,it is not hard to assigns equal mass to the components of the cycle.
80 L. M. BERLINER
Note that all these distributionsare mutuallyor- the stronglaw of large numbers.Pursuing the
thogonal. An implicationof these considerations deterministic argument,I will leave it to philoso-
suggest that, to observe ergodic behavior corre- phersto debatethemeaningofthe claim,"Starting
spondingto the arc sine forthe logisticwith a = 4, at initial conditionxo = .2958672..., the proba-
we mustparticularlyavoid the preimages(see Sec- bility that Yn = 1 and Yn+1 = 0 is 0.52 if
tion2) ofall periodicpoints. n = 101oo.''Althoughthere is nothing"random"
Actually,the above argumentsofferonly a par- statementstillmakessense to
here,the probability
tial explanationofwhywe see ergodic,apparently me.
randombehaviorin computer-generated dynamical
systems.Electroniccomputerscan onlyrepresenta 3. DATA ANALYSIS AND CHAOS
finite,although,fortunately, quite large numberof
In thissectionI will reviewsomeofthe problems
numbersand thus no numericallycomputeddy-
associatedwithchaoticmodelsand data thatare of
namicalsystemcan trulybe aperiodic.In the logis-
particularto statisticians.
tic map exampleofFigure 8, the initial conditions
were necessarilyonlytruncatedreal numbers,and 3.1 Measuring Chaos
thus, lie in a set of Lebesgue measure zero. A
possible explanationof why we still can observe Importantquestionsarise in attempting to char-
behavior expectedunder the ergodictheoremin- acterizewhat a chaotictime series of data should
volves the so-called shadowing property.The idea look like. Intuitively,a chaoticseries should look
is that the computer-generated orbitofthe system "random,"butthisintuitionis notnecessarilyeasy
is, in a sense,an approximation to sometrueorbit. to quantifyin termsofthe mathematicalor proba-
The result is that we can be reasonablyconfident discussedin Section2.
bilisticdefinitions
that,ifpropercare is taken,computerresults,espe- 3.1.1 Liapunov exponents
cially aggregatedresultssuch as the histogramsof
Figure 8, do in factcapturethe correctqualitative One of the most popular measures of chaos,
featuresofthe system.The requiredcare alludedto Liapunov exponents,are based on mathematics as-
in the previous statementrefersto the roundoff sociated with the sensitivityto initial conditions
errorpresentin the numericalcomputationofthe conceptdescribedin Section 2. [Nearly all of the
nonlinearfunctionf. Furtherdiscussionofcompu- general referencesgiven here presentdiscussions;
tationalissues fordynamicalsystemsis beyondthe especiallysee Eckmann and Ruelle (1985).] Con-
scope ofthis paper [see Guckenheimer and Holmes sidera univariate,discretetimedynamicalsystem
(1983); Hammel, Yorke and Grebogi(1987); and where xn fn( xo). To studythe impact of varying
Parkerand Chua (1989)]. initial conditions,it is natural to considerderiva-
The storyof ergodictheory,chaos and random- tives dxn/dxo. The Liapunov exponent,say X(so),
ness is stillnotcomplete.A mostintimaterelation- is definedas (x0) = limn- (1/n) log[Idxn/dxo ].
ship accrues fromthe followingresult. Under the Notethatvia the chainrule,dxn/dx0can be repre-
assumptionsof the ergodictheorem,if the initial sentedas a productand so X((xo), underappropriate
conditionxo is generatedaccordingto P, the se- conditions,may be subjectto the ergodictheorem.
quence { Y(fi(xo)) = Yi(xo), i > 0} is a stationary If so, then X(x0)= X, independentof x0, almost
stochasticprocess. (The probabilisticstructureof surelywithrespectto an appropriateergodicdistri-
this stochasticprocessvaries withthe ergodicdis- bution.Under such circumstancesX is a quantita-
tributionused to generatexo.) Thus,ifwe chooseY tive measure of the dynamicalsystem'sdegreeof
to be the identitytransformation, Y(x) = x, the sensitivityto initial conditions.In particular,the
deterministic dynamicalsystemwithinitial condi- approximation
tion somehowchosen,with care to avoid sets of (3.1) dxn z eXndxo
P-measurezero,is a realizationofa stochasticproc-
ess. This result is also importantin symbolicdy- suggeststhat forlarge X,small changesin initial
namics;indeed,the previousstatementprovidesa conditionsresultin separationofpaths at an expo-
naturalgeneralizeddefinition ofchaos in the spirit nentialrate as n grows.
of Definition2.4. Considerthe symbolicdefinition Notethat,in the ergodiccase, Xcan be estimated
ofSection2.3 forthe logisticwhen a = 4 and S = froma singletimeseries.This meanswe can assess
[O,.5].In this case, if the initial conditionhas the sensitivityto initial conditionseven thoughthe
arc sine distribution, it can be shownthat the Y's data is based on a single x0. Of course,this as-
are actually independent,identicallydistributed sumes that ergodicity applies and thatthe xo that
("equally likely")Bernoullirandomvariables.[For generatedthe data is in supportofthe appropriate
related discussion,see Breiman (1968, page 108).] ergodicdistribution.(Recall the ergodicdistribu-
In such a case, the ergodictheoremcoincideswith tions, althoughorthogonal,need not be unique.)
CHAOS 81
15t X2nd\ X
R 10 R 10
E E
T T0 \
U 0- U O
RU R
-10 -10
-10 0 10 -10 0 10
I I
3rd loth
R 10 - R 10
I! E
T )T
U 0 *,, U 0
RR
NN
-' - '--;., ',il
-10-\ ' -IO
-10 0 10 -10 0 10
ia}
I o.
... ...
..
.........
.
.. .. . '''. ' .. ..'
-10- .. ,..:....'
tweenobservationtimesand short-time prediction. cal regularityby Hopf.See Engel (1987) fora very
The potentialhere is actuallyverylarge. I cannot valuable discussion.To quicklycommunicatethe
dojusticeto the directionhere,but offerthe follow- idea, recall the ergodictheoryreview of Section
ing referencesto the interestedreader: Lumley 2.2.3 and the corresponding exampleofthe logistic
(1970), Crutchfleldand McNamara (1987), Farmer map with a = 4. In this case the arc-sinedistri-
and Sidorowich (1987), Casdagli (1989), and bution is a continuous, ergodic distribution.
Kostelich and Yorke (1990). Furthermore,the Invarianceimpliesthat if xo is generatedaccord-
referenceby Nychka,McCaffrey, Ellner and Gal- ingto the arc-sine,thenat everytime t, xt also has
lant (1990) is especiallyrecommendedto statisti- an arc-sinedistribution.However,some intuition
cians forits new results,as well as reviewin this suggestedthat xo need not actuallybe generated
direction. accordingto the ergodicdistributionfor ergodic
behaviorto manifestitself.Hopf'snotionofstatisti-
4. STATISTICAL ANALYSES FOR CHAOS cal regularityis a rigorousresultalongtheselines.
The result is that, under mild regularitycondi-
4.1 ParametricStatistical Analysis forChaotic tions,includinga continuousergodicdistribution,
Models say P, if xo has any distributionthat is absolutely
Geweke (1989) and Berliner(1991) are the pri- continuouswith respectto P, then as t tends to
maryreferences forthis section.A naturalclass of infinity,xt convergesto distributionto P. That is,
modelsforwhichthe statisticianfeels "at home" the initial distribution"washes out." (Note the
are based on the specification
ofa dynamicalfunc- natural correspondence between this notion and
tion drivingthe systemunderstudy.Assume that the conceptof stationary,ergodicdistributions in
the functionis knownup to a finitecollectionof Markov processes.) The applicationto Bayesian
parameters.Specifically,the dynamicalsystemis forecastingis immediate.Under the appropriate
assumedto be drivenby the relationship conditions,if we computea posteriordistribution
forunknowninitialconditionsand thatposterioris
(4.1) xt+1= f(xt; -0 abso-lutelycontinuouswithrespectto a continuous
ergodic distribution, then our implied predictive
where f is specifled.The parameter - and the distribution forxt as t growsmustcollapseto the
initial value xo may both be unknown.Further, ergodicdistribution.This is a strong,statistical
assume that,at some time points,we observethe reflectionofthenotionofunpredictability ofchaotic
x-processwithobservationerror.Dependingon the processes.The strength oftheresultis thatthereis
model we use for the errors,we an constructa nothingphilosophicalto debate beforeaccepting
likelihoodfunctionbased on data forthe unknown implicationto forecasting.The theoremsays that
quantities.Even in verysimpleexamples,such as underperfect conditionsin whichtheprioris agreed
the logisticmap observedwithindependentGauss- to be known and correctlyspecified,and thus,
ian errors,the resultinglikelihoodfunctionscan be Bayesian computationsare uncontroversial appli-
extremelycomplex,intractableobjects. Berliner cationsofprobability theory,long-term predictions,
(1991) offerssomeheuristicsconcerning the behav- moreprecisethan thoseassociatedwitha continu-
ior ofsuch "chaoticlikelihoods."For example,it is ous ergodicdistribution, of chaotic processesare
possibleto relate chaos as measuredvia Liapunov impossible.
exponentswithFisher information concerningun-
known initial conditions.Chaotic processes ob- 4.2 Statistics and Dynamical Systems
served with errorproducestatisticalinformation There is a huge literaturedevotedto statistical
concerning initialconditions.However,thevalue of analysesfordynamicalsystems.Statisticiansregu-
this information forpredictionis limited.Specifi- larlyconsiderthe modelwith"systemequation"
cally, if maximumlikelihoodestimatesof initial
conditionsare sought,one is firstfacedwitha very (4.2) xt+1 = f(xt;') + zt,
difflcultproblemoffindingsuch estimates.Even if
one were able to obtain a good estimate of xo, where { zt} is itselfa stochasticprocess,and "ob-
sensitivityto initialconditionsmoderatesthe value servationequation"forthe observabley
ofsuch estimatesin the contextofprediction.
(4.3) Yt = h( xt) + et,
Berliner (1991) considersBayesian forecasting
based on modelsas suggestedin the previouspara- where h is some functionand e representsmeas-
graph. Bayesian forecasting,in the presence of urementerror.The inclusionof z in (4.2) is sug-
unknowninitial conditions,is intimatelyrelated gestedas a natural,morerealisticversionof(4.1),
to ergodictheoryvia a phenomenoncalled statisti- whichallows some notionoferrorin the specifica-
CHAOS 87