Searcharxiv⌕ Search

arXiv subjects

D. Sornette

Publications and source records attributed to D. Sornette.

At least 217 records · Page 12Linked to original sources
arXiv2000

Re-examination of log-periodicity observed in the seismic precursors of the 1989 Loma Prieta earthquake

Based on several empirical evidence, a series of papers has advocated the concept that seismicity prior to a large earthquake can be understood in terms of the statistical physics of a critical phase transition. In this model, the cumulative Benioff strain (BS) increases as a power-law time-to-failure before the final event. This power law reflects a kind of scale invariance with respect to the distance to the critical point. A few years ago, on the basis of a fit of the cumulative BS released prior to the 1989 Loma Prieta earthquake, Sornette and Sammis [1995] proposed that this scale invariance could be partially broken into a discrete scale invariance (DSI). The observable consequence of DSI takes the form of log-periodic oscillations decorating the accelerating power law. They found that the quality of the fit and the predicted time of the event are significantly improved by the introduction of log-periodicity. Here, we present a battery of synthetic tests performed to quantify the statistical significance of this claim. We find that log-periodic oscillations with frequency and regularity similar to those of the Loma Prieta case are very likely to be generated by the interplay of the low pass filtering step due to the construction of cumulative functions together with the approximate power law acceleration. Thus, the single Loma Prieta case alone cannot support the initial claim and additional cases and further study are needed to increase the signal-to-noise ratio if any. The present study will be a useful methodological benchmark for future testing of additional events when the methodology and data to construct reliable Benioff strain function become available.

physics.geo-ph↗
arXiv2000

Stock Market Speculation: Spontaneous Symmetry Breaking of Economic Valuation

Firm foundation theory estimates a security's firm fundamental value based on four determinants: expected growth rate, expected dividend payout, the market interest rate and the degree of risk. In contrast, other views of decision-making in the stock market, using alternatives such as human psychology and behavior, bounded rationality, agent-based modeling and evolutionary game theory, expound that speculative and crowd behavior of investors may play a major role in shaping market prices. Here, we propose that the two views refer to two classes of companies connected through a ``phase transition''. Our theory is based on 1) the identification of the fundamental parity symmetry of prices ($p \to -p$), which results from the relative direction of payment flux compared to commodity flux and 2) the observation that a company's risk-adjusted growth rate discounted by the market interest rate behaves as a control parameter for the observable price. We find a critical value of this control parameter at which a spontaneous symmetry-breaking of prices occurs, leading to a spontaneous valuation in absence of earnings, similarly to the emergence of a spontaneous magnetization in Ising models in absence of a magnetic field. The low growth rate phase is described by the firm foundation theory while the large growth rate phase is the regime of speculation and crowd behavior. In practice, while large ``finite-time horizon'' effects round off the predicted singularities, our symmetry-breaking speculation theory accounts for the apparent over-pricing and the high volatility of fast growing companies on the stock markets.

cond-mat.stat-mech↗
arXiv2000

Economy of scales in R&D with block-busters

Are large scale research programs that include many projects more productive than smaller ones with fewer projects? This problem of economy of scale is particularly relevant for understanding recent mergers in particular in the pharmaceutical industry. We present a quantitative theory based on the characterization of distributions of discounted sales S resulting from new products. Assuming that these complementary cumulative distributions have fat tails with approximate power law structure S^{-a}, we demonstrate that economy of scales are automatically realized when the exponent a is less than one. Empirical evidence suggests that the exponent a is approximately equal to 2/3 for the pharmaceutical industry.

cond-mat.stat-mech↗
arXiv2000

The critical earthquake concept applied to mine rockbursts with time-to-failure analysis

We report new tests of the critical earthquake concepts performed on rockbursts in deep South African mines. We extend the concept of an optimal time and space correlation region and test it on the eight main shocks of our catalog provided by ISSI. In a first test, we use the simplest signature of criticality in terms of a power law time-to-failure formula. Notwithstanding the fact that the search for the optimal correlation size is performed with this simple power law, we find evidence both for accelerated seismicity and for the presence of logperiodic behavior with a prefered scaling factor close to 2. We then propose a new algorithm based on a space and time smoothing procedure, which is also intended to account for the finite range and time mechanical interactions between events. This new algorithm provides a much more robust and efficient construction of the optimal correlation region, which allows us the use of the logperiodic formula directly in the search process. In this preliminary work, we have only tested the new algorithm on the largest event on the catalog. The result is of remarkable good quality with a dramatic improvement in accuracy and robustness. This confirms the potential importance of logperiodic signals. Our study opens the road for an efficient implemention of a systematic testing procedure of real-time predictions.

cond-mat.stat-mech↗
arXiv2000

"Thermometers" of Speculative Frenzy

Establishing unambiguously the existence of speculative bubbles is an on-going controversy complicated by the need of defining a model of fundamental prices. Here, we present a novel empirical method which bypasses all the difficulties of the previous approaches by monitoring external indicators of an anomalously growing interest in the public at times of bubbles. From the definition of a bubble as a self-fulfilling reinforcing price change, we identify indicators of a possible self-reinforcing imitation between agents in the market. We show that during the build-up phase of a bubble, there is a growing interest in the public for the commodity in question, whether it consists in stocks, diamonds or coins. That interest can be estimated through different indicators: increase in the number of books published on the topic, increase in the subscriptions to specialized journals. Moreover, the well-known empirical rule according to which the volume of sales is growing during a bull market finds a natural interpretation in this framework: sales increases in fact reveal and pinpoint the progress of the bubble's diffusion throughout society. We also present a simple model of rational expectation which maps exactly onto the Ising model on a random graph. The indicators are then interpreted as ``thermometers'', measuring the balance between idiosyncratic information (noise temperature) and imitation (coupling) strength. In this context, bubbles are interpreted as low or critical temperature phases, where the imitation strength carries market prices up essentially independently of fundamentals. Contrary to the naive conception of a bubble and a crash as times of disorder, on the contrary, we show that bubbles and crashes are times where the concensus is too strong.

cond-mat.stat-mech↗
arXiv2000

Acoustic fluidization for earthquakes?

Melosh [1996] has suggested that acoustic fluidization could provide an alternative to theories that are invoked as explanations for why some crustal faults appear to be weak. We show that there is a subtle but profound inconsistency in the theory that unfortunately invalidates the results. We propose possible remedies but must acknowledge that the relevance of acoustic fluidization remains an open question.

cond-mat.stat-mech↗
arXiv1999

New Evidence of Earthquake Precursory Phenomena in the 17 Jan. 1995 Kobe Earthquake, Japan

Significant advances, both in the theoretical understanding of rupture processes in heterogeneous media and in the methodology for characterizing critical behavior, allows us to reanalyze the evidence for criticality and especially log-periodicity in the previously reported chemical anomalies that preceded the Kobe earthquake. The ion ($Cl^-$, $K^+$, $Mg^{++}$, $NO_3^{-}$ and $SO_4^{--}$) concentrations of ground-water issued from deep wells located near the epicenter of the 1995 Kobe earthquake are taken as proxies for the cumulative damage preceding the earthquake. Using both a parametric and non-parametric analysis, the five data sets are compared extensively to synthetic time series. The null-hypothesis that the patterns documented on these times series result from noise decorating a simple power law is rejected with a very high confidence level.

cond-mat↗
arXiv1999

On Rational Bubbles and Fat Tails

This paper addresses the statistical properties of time series driven by rational bubbles a la Blanchard and Watson (1982), corresponding to multiplicative maps, whose study has recently be revived recently in physics as a mechanism of intermittent dynamics generating power law distributions. Using insights on the behavior of multiplicative stochastic processes, we demonstrate that the tails of the unconditional distribution emerging from such bubble processes follow power-laws (exhibit hyperbolic decline). More precisely, we find that rational bubbles predict a 'fat' power tail for both the bubble component and price differences with an exponent smaller than 1, implying absence of convergence of the mean. The distribution of returns is dominated by the same power-law over an extended range of large returns. Although power-law tails are a pervasive feature of empirical data, these numerical predictions are in disagreement with the usual empirical estimates of an exponent between 2 and 4. It, therefore, appears that exogenous rational bubbles are hardly reconcilable with some of the stylized facts of financial data at a very elementary level.

cond-mat.stat-mech↗
arXiv1999

Fundamental Framework for Technical Analysis

Starting from the characterization of the past time evolution of market prices in terms of two fundamental indicators, price velocity and price acceleration, we construct a general classification of the possible patterns characterizing the deviation or defects from the random walk market state and its time-translational invariant properties. The classification relies on two dimensionless parameters, the Froude number characterizing the relative strength of the acceleration with respect to the velocity and the time horizon forecast dimensionalized to the training period. Trend-following and contrarian patterns are found to coexist and depend on the dimensionless time horizon. The classification is based on the symmetry requirements of invariance with respect to change of price units and of functional scale-invariance in the space of scenarii. This ``renormalized scenario'' approach is fundamentally probabilistic in nature and exemplifies the view that multiple competing scenarii have to be taken into account for the same past history. Empirical tests are performed on on about nine to thirty years of daily returns of twelve data sets comprising some major indices (Dow Jones, SP500, Nasdaq, DAX, FTSE, Nikkei), some major bonds (JGB, TYX) and some major currencies against the US dollar (GBP, CHF, DEM, JPY). Our ``renormalized scenario'' exhibits statistically significant predictive power in essentially all market phases. In constrast, a trend following strategy and trend + acceleration following strategy perform well only on different and specific market phases. The value of the ``renormalized scenario'' approach lies in the fact that it always finds the best of the two, based on a calculation of the stability of their predicted market trajectories.

cond-mat.stat-mech↗
arXiv1999

Market Fluctuations: multiplicative and percolation models, size effects and predictions

We present a set of models of the main stylized facts of market price fluctuations. These models comprise dynamical evolution with threshold dynamics and Langevin price equation with multiplicative noise, percolation models to describe the interaction between traders and hierarchical cascade models to unravel the possible correlation accross time scales, including the log-periodic signatures associated to financial crashes. The main empirical knowledge is summarized and some key empirical tests are presented.

cond-mat.stat-mech↗
arXiv1999

Have your cake and eat it too: increasing returns while lowering large risks!

Based on a faithful representation of the heavy tail multivariate distribution of asset returns introduced previously (Sornette et al., 1998, 1999) that we extend to the case of asymmetric return distributions, we generalize the return-risk efficient frontier concept to incorporate the dimensions of large risks embedded in the tail of the asset distributions. We demonstrate that it is often possible to increase the portfolio return while decreasing the large risks as quantified by the fourth and higher order cumulants. Exact theoretical formulas are validated by empirical tests.

cond-mat.stat-mech↗
arXiv1999

Self-Organized Percolation Model for Stock Market Fluctuations

In the Cont-Bouchaud model [cond-mat/9712318] of stock markets, percolation clusters act as buying or selling investors and their statistics controls that of the price variations. Rather than fixing the concentration controlling each cluster connectivity artificially at or close to the critical value, we propose that clusters shatter and aggregate continuously as the concentration evolves randomly, reflecting the incessant time evolution of groups of opinions and market moods. By the mechanism of ``sweeping of an instability'' [D. Sornette, Journal de Physique I 4, 209 (1994)], this market model spontaneously exhibits reasonable power law statistics for the distribution of price changes and accounts for the other important stylized facts of stock market price fluctuations.

cond-mat.stat-mech↗
arXiv1999

Analysis of the phenomenon of speculative trading in one of its basic manifestations: postage stamp bubbles

We document and analyze the empirical facts concerning one of the clearest evidence of speculation in financial trading as observed in the postage collection stamp market. We unravel some of the mechanisms of speculative behavior which emphasize the role of fancy and collective behavior. In our conclusion, we propose a classification of speculative markets based on two parameters, namely the amplitude of the price peak and a second parameter that measures its ``sharpness''. This study is offered to anchor modeling efforts to realistic market constraints and observations.

cond-mat.stat-mech↗
arXiv1999

Renormalization of earthquake aftershocks

Together with the Gutenberg-Richter distribution of earthquake magnitudes, Omori's law is the best established empirical characterization of earthquake sequences and states that the number of smaller earthquakes per unit time triggered by a main shock decays approximately as the inverse of the time ($1/t^p$, with $p \approx 1$) since the main shock. Based on these observations, we explore the theoretical hypothesis in which each earthquake can produce a series of aftershock independently of its size according to its ``local'' Omori's law with exponent $p=1+θ$. In this scenario, an aftershock of the main shock produces itself other aftershocks which themselves produce aftershocks, and so on. The global observable Omori's law is found to have two distinct power law regimes, the first one with exponent $p_-=1 - θ$ for time $t < t^* \sim κ^{-1/θ}$, where $0<1-κ<1$ measures the fraction of triggered earthquakes per triggering earthquake, and the second one with exponent $p_+=1 + θ$ for larger times. The existence of these two regimes rationalizes the observation of Kisslinger and Jones [1991] that the Omori's exponent $p$ seems positively correlated to the surface heat flow: a higher heat flow is a signature of a higher crustal temperature, which leads to larger strain relaxation by creep, corresponding to fewer events triggered per earthquake, i.e. to a larger $κ$, and thus to a smaller $t^*$, leading to an effective measured exponent more heavily weighted toward $p_+>1$.

cond-mat.stat-mech↗
arXiv1999

General theory of the modified Gutenberg-Richter law for large seismic moments

The Gutenberg-Richter power law distribution of earthquake sizes is one of the most famous example illustrating self-similarity. It is well-known that the Gutenberg-Richter distribution has to be modified for large seismic moments, due to energy conservation and geometrical reasons. Several models have been proposed, either in terms of a second power law with a larger b-value beyond a cross-over magnitude, or based on a ``hard'' magnitude cut-off or a ``soft'' magnitude cut-off using an exponential taper. Since the large scale tectonic deformation is dominated by the very largest earthquakes and since their impact on loss of life and properties is huge, it is of great importance to constrain as much as possible the shape of their distribution. We present a simple and powerful probabilistic theoretical approach that shows that the Gamma distribution is the best model, under the two hypothesis that the Gutenberg-Richter power law distribution holds in absence of any condition (condition of criticality) and that one or several constraints are imposed, either based on conservation laws or on the nature of the observations themselves. The selection of the Gamma distribution does not depend on the specific nature of the constraint. We illustrate the approach with two constraints, the existence of a finite moment release rate and the observation of the size of a maximum earthquake in a finite catalog. Our predicted ``soft'' maximum magnitudes compare favorably with those obtained by Kagan [1997] for the Flinn-Engdahl regionalization of subduction zones, collision zones and mid-ocean ridges.

cond-mat.stat-mech↗
arXiv1999

Persistence and Quiescence of Seismicity on Fault Systems

We study the statistics of simulated earthquakes in a quasistatic model of two parallel heterogeneous faults within a slowly driven elastic tectonic plate. The probability that one fault remains dormant while the other is active for a time Dt following the previous activity shift is proportional to the inverse of Dt to the power 1+x, a result that is robust in the presence of annealed noise and strength weakening. A mean field theory accounts for the observed dependence of the persistence exponent x as a function of heterogeneity and distance between faults. These results continue to hold if the number of competing faults is increased. This is related to the persistence phenomenon discovered in a large variety of systems, which specifies how long a relaxing dynamical system remains in a neighborhood of its initial configuration. Our persistence exponent is found to vary as a function of heterogeneity and distance between faults, thus defining a novel universality class.

cond-mat.stat-mech↗
arXiv1999

Novel Mechanism for Discrete Scale Invariance in Sandpile Models

Numerical simulations and a mean-field analysis of a sandpile model of earthquake aftershocks in 1d, 2d and 3d euclidean lattices determine that the average stress decays in a punctuated fashion after a main shock, with events occurring at characteristic times increasing as a geometrical series with a well-defined multiplicative factor which is a function of the stress corrosion exponent, the stress drop ratio and the degree of dissipation. These results are independent of the discrete nature of the lattice and stem from the interplay between the threshold dynamics and the power law stress relaxation.

cond-mat.stat-mech↗