Several concepts of fractal dimension have been developed to characterise properties of attractors of chaotic dynamical systems. Numerical approximations of them must be calculated by finite samples of simulated trajectories. In principle, the quantities should not depend on the choice of the trajectory, as long as it provides properly distributed samples of the underlying attractor. In practice, however, the trajectories are sensitive with respect to varying initial values, small changes of the model parameters, to the choice of a solver, numeric tolerances, etc. The purpose of this paper is to present a statistically sound approach to quantify this variability. We modify the concept of correlation integral to produce a vector that summarises the variability at all selected scales. The distribution of this stochastic vector can be estimated, and it provides a statistical distance concept between trajectories. Here, we demonstrate the use of the distance for the purpose of estimating model parameters of a chaotic dynamic model. The methodology is illustrated using computational examples for the Lorenz 63 and Lorenz 95 systems, together with a framework for Markov chain Monte Carlo sampling to produce posterior distributions of model parameters.

1.
H.
Haario
,
E.
Saksman
, and
J.
Tamminen
,
Bernoulli
7
,
223
(
2001
).
2.
H.
Haario
,
M.
Laine
,
A.
Mira
, and
E.
Saksman
,
Stat. Comput.
16
,
339
(
2006
).
3.
G.
Boffetta
,
A.
Celani
,
M.
Cencini
,
G.
Lacorata
, and
A.
Vulpiani
,
J. Phys. A: Math. Gen.
33
,
1313
(
2000
).
4.
C.
Robert
and
G.
Casella
,
Monte Carlo Statistical Methods
, 2nd ed. (
Springer
,
2004
).
5.
B. E.
Kendall
,
C. J.
Briggs
,
W. W.
Murdoch
,
P.
Turchin
,
S. P.
Ellner
,
E.
McCauley
,
R. M.
Nisbet
, and
S. N.
Wood
,
Ecology
80
,
1789
(
1999
).
6.
M. A.
Beaumont
,
W.
Zhang
, and
D. J.
Balding
,
Genetics
162
,
2025
(
2002
); available at http://www.ncbi.nlm.nih.gov/pubmed/12524368.
7.
D.
MacFadden
,
Econometrica
57
,
995
(
1989
).
9.
J.
Rougier
,
Philos. Trans. R. Soc. London, Ser. A
371
,
20120297
(
2013
).
10.
H.
Järvinen
,
P.
Räisänen
,
M.
Laine
,
J.
Tamminen
,
J. A.
Ilin
,
E.
Oja
,
A.
Solonen
, and
H.
Haario
,
Atmos. Chem. Phys.
10
,
9993
(
2010
).
11.
J.
Hakkarainen
,
A.
Ilin
,
A.
Solonen
,
M.
Laine
,
H.
Haario
,
J.
Tamminen
,
E.
Oja
, and
H.
Järvinen
,
Nonlinear Processes Geophys.
19
,
127
(
2012
).
12.
J.
Hakkarainen
,
A.
Solonen
,
A.
Ilin
,
J.
Susiluoto
,
M.
Laine
,
H.
Haario
, and
H.
Järvinen
,
Tellus Ser. A
65
,
20147
(
2013
).
13.
G.
Malinetskii
and
A.
Potapov
,
Contemporary Problems of Nonlinear Dynamics
, 2nd ed. (
URSS
,
Moscow
,
2002
) (in Russian).
14.
P.
Grassberger
and
I.
Procaccia
,
Phys. Rev. A
28
,
2591
(
1983
).
15.
P.
Grassberger
and
I.
Procaccia
,
Physica D
9
,
189
(
1983
).
16.
P.
Grassberger
,
Phys. Lett. A
148
,
63
(
1990
).
17.
G.
Malinetskii
and
A.
Potapov
,
USSR Comput. Math. Math. Phys.
28
,
39
(
1988
).
18.
J.
Theiler
, “
Estimating fractal dimension
,”
J. Opt. Soc. Am. A
7
,
1055
(
1990
).
19.
A.
Potapov
and
J.
Kurths
,
Physica D
120
,
369
(
1998
).
20.
R.
Field
,
P.
Hess
,
L.
Kalachev
, and
S.
Madronich
,
J. Geophys. Res.
106
,
7552
, doi: (
2001
).
21.
R. M.
Dudley
,
Uniform Central Limit Theorems
(
Cambridge University Press
,
New York
,
1999
).
22.
F.
Cecconi
,
M.
Cencini
,
M.
Falcioni
, and
A.
Vulpiani
,
Am. J. Phys.
80
,
1001
(
2012
).
23.
D.
Ruelle
,
Proc. R. Soc. London, Ser. A
427
,
1873
(
1990
).
24.
H.
Kantz
and
T.
Schreiber
,
Nonlinear Time Series Analysis
(
Cambridge University Press
,
Cambridge, UK
,
1997
).
25.
A. H.
Nayfeh
and
B.
Balachandra
,
Applied Nonlinear Dynamics
(
John Wiley
,
New York
,
1995
).
26.
T. S.
Parker
and
L. O.
Chua
,
Practical Numerical Algorithms for Chaotic Systems
(
Springer-Verlag
,
New York
,
1989
).
27.
28.
C.
Andrieu
and
G. O.
Roberts
,
Ann. Stat.
37
,
697
(
2009
).
29.
C.
Andrieu
and
M.
Vihola
,
Ann. Appl. Probab.
(to be published); preprint arXiv:1210.1484v2 (
2014
).
30.
M.
Beaumont
,
Genetics
164
,
1139
(
2003
); available at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1462617/.
You do not currently have access to this content.