Random Walk Behavior Research Articles

Learning integrated inflation forecasts in a simple multi-agent macroeconomic model

This paper implements a model with a population of heterogeneous macro forecasters. Their objectives are to forecast output and inflation, both inputs in standard New Keynesian macro models. The model is implemented by first calibrating the agents to professional forecasters at the micro level. Model runs then try to replicate both the dynamics, bias and cross sectional heterogeneity of forecasts and the economy. These are done both in a model with static forecasters, and one where the forecasters are learning from each other in a social fashion. We find that expectations about the inflation process which conjecture near random walk behavior can be self-fulfilling, yielding inflation volatility and persistence on the order of magnitude of U.S. macro data. However, our forecasting populations often fall short of the heterogeneity of predictions from survey data. In some cases, monetary policy can be used to shift the model from its volatile/persistent equilibrium over to a more stable, strongly mean reverting inflation rate.

A Kalman-filter based strategy for space-frequency force reconstruction

This contribution presents a Kalman filtering strategy for space-frequency reconstruction of mechanical sources. The proposed approach is based on the definition of an appropriate state-space model, where the state equation assumes that the force vector follows a random walk behavior, while the observation equation is the classical linear relationship between the force and the measured response through the transfer function matrix of the considered structure. One of the main challenges of the proposed approach is the fine-tuning of the covariance matrix associated with the process noise. In this work, it is estimated and adapted at each frequency during the filtering procedure. A numerical experiment is performed on a simply supported beam excited by a broadband point mechanical force to evaluate the reconstruction performance of the proposed approach. Further comparisons with other regularization strategies are also proposed to provide a fair overview of the results obtained.

Multifractal Analysis of the Impact of Fuel Cell Introduction in the Korean Electricity Market

This study employs multifractal detrended fluctuation analysis to investigate the impact of fuel cell introduction in the Korean electricity market via the lens of multifractal scaling behavior. Using multifractal analysis, the research delineates discrepancies between peak and off-peak hours, accounting for the daily cyclicity of the electricity market, and proposes a crossover point detection method based on the Chow test. Furthermore, the impacts of fuel cell introduction are evidenced through various methods that encompass multifractal spectra and market efficiency. The findings initially indicate a higher degree of multifractality during off-peak hours relative to peak hours. In particular, the crossover points emerged solely during off-peak hours, unveiling short- and long-term dynamics predicated on a near-annual cycle. Additionally, the average Hurst exponent for the short-term was 0.542, while the average for the long-term was 0.098, representing a notable discrepancy. The introduction of fuel cells attenuated the heterogeneity in the scaling behavior, which is potentially attributable to the decreased volatility in both the supply and demand spectra. Remarkably, after the introduction of fuel cells, there was a discernible decrease in the influence of long-range correlation within multifractality, and the market exhibited an increased propensity toward random-walk behavior. This phenomenon was also detected in the market deficiency measure, from an average of 0.536, prior to the introduction, to an average of 0.267, following the introduction, signifying an improvement in market efficiency. This implies that the introduction of fuel cells into the market engendered increased supply stability and a consistent increase in demand, mitigating volatility on both the supply and demand sides, thus increasing market efficiency.

Using dynamic semantic structure of news flow to enhance financial forecasting: a twelve-year study on twitter news channels

AbstractThis research holds significance for advancing financial forecasting methodologies by shifting the focus from traditional sentiment analysis of individual tweets to exploring intricate semantic relationships within news tweets from top-followed news channels on Twitter. Addressing a notable research gap in financial forecasting, often dominated by sentiment analysis, our study endeavors to fill the void left by the underexplored intricate relationships within news entities and their dynamic semantic evolution. Motivated by the inherent challenges in predicting the random walk behavior of stock prices, we contend that incorporating longitudinal data derived from the semantic relationships between news entities can enhance the accuracy of stock market forecasts. The study pioneers a twelve-year exploration, encompassing data from 55 leading news channels on Twitter, boasting a collective following of 714 million users. The approach employs natural language processing (NLP) to extract two million unique entities, whose semantics are analyzed through complex network analysis, laying the foundation for the forecasting model. Finally, this research introduces a model linked to the dynamic semantic structure of news flow. The predictive model considers the impact of exogenous variables influenced by the evolving relationships among news entities. The results offer a proof of concept, highlighting the potential of utilizing dynamic semantic relationships among news entities for financial prediction. On average, the model demonstrates an improvement in accuracy of 40.3% across ten different stock price predictions. These findings are expounded through relevant theories, offering a theoretical foundation for observed patterns and indicating a promising direction for future research in this domain.

Chaotic and thermal aspects in the highly excited string S-matrix

We compute tree level scattering amplitudes involving more than one highly excited states and tachyons in bosonic string theory. We use these amplitudes to understand the chaotic and thermal aspects of the excited string states lending support to the Susskind-Horowitz-Polchinski correspondence principle. The unaveraged amplitudes exhibit chaos in the resonance distribution as a function of the kinematic parameters, which can be described by random matrix theory. Upon coarse-graining, these amplitudes are shown to exponentiate, and capture various thermal features, including features of a stringy version of the eigenstate thermalization hypothesis as well as notions of typicality. Further, we compute the effective string form factor corresponding to the highly excited states, and argue for the random walk behaviour of the long strings.

Adaptive Controller Based on Estimated Parameters for Quadcopter Trajectory Tracking

This paper presents a trajectory control system design for a quadcopter, an unmanned aerial vehicle (UAV), which is based on estimated parameters that are assumed to exhibit random walk behavior. Initially, the rotational dynamic model of the UAV is formulated using the Newton Euler method in terms of angular velocity about the x, y, and z axes. This model is then simplified into three separated-first-order linear differential equations, with coefficients derived from the combined effects of inertia, aerodynamic drag, and gyroscopic effects, referred to as lumped parameters. A Proportional-Integral (PI) controller with feed-forward design is then developed to control this simplified model. To adapt the controller to the lumped parameters that exhibit random walk behavior, each simplified equation is restructured into a processing and measurement model. The states of these models are estimated by using the Unscented Kalman Filter (UKF). These estimated values are then utilized to adjust the PI gains and compensate the signal of the designed angular velocity controller, transforming it into an adaptive controller. The entire UAV controller comprises two main parts, an inner loop for adaptive angular rate control and an outer loop serving as an attitude-thrust controller. The proposed controller is simulated using Simulink, with circular and square trajectories. The simulation results demonstrate that the quadcopter successfully follows the desired circular and square paths. The steady-state error for the x and y axes in the square trajectory is less than 0.05 meters within 5 seconds, and for the z axis, it is less than 0.02 meters within 2.5 seconds. The controller gains do not require adjustment when changing trajectories. Moreover, the estimated parameters remain nearly constant at steady state.

Random walk tests for the MENA stock returns

Purpose — The current study seeks to understand whether individual stock returns exhibit random movement and are not dependent (efficient at weak form) on fourteen out of sixteen actively traded Arab stock markets in the Middle East and North Africa (MENA) region, based on the size of the market value.Method — Various non-parametric methods, including autocorrelation test, variance ratio test, Phillips-Perron unit root test, and runs test, are used to assess the random walk hypothesis for daily data following the Covid-19 vaccination program. This analysis covers the period from January 3, 2021, to March 28, 2023.Result — The study results present evidence that all individual stock returns deviate from random walk behavior. However, only Kuwait, Jordan, and Palestine stock returns follow the random walk based on the run test results at a significance level of 10%. Therefore, it can be concluded that all stock returns are inefficient at the weak-form, suggesting that investors have opportunities for unexpected gains.Practical implications — The findings of this study suggest that investors in the MENA region may have opportunities for unexpected gains, as individual stock returns deviate from random walk behavior, highlighting the importance of considering market dynamics and employing informed investment strategies. Additionally, policymakers could benefit from understanding the inefficiencies in stock returns to implement measures that promote market stability and efficiency.

TESTING WEAK-FORM EFFICIENCY FOR THE SAUDI PARALLEL MARKET (NOMU)

Market efficiency is essential for a thriving stock exchange and has a substantial impact on capital investments, particularly in emerging markets. This article examines the weak-form efficiency of Nomu, the Saudi Arabian parallel stock exchange. The study employed various statistical tests on the collected closing values of the Nomu index from May 1, 2019, to May 31, 2023. The return normality and stationarity were investigated using the Jarque-Bera, Augmented Dickey-Fuller (ADF), and Phillips-Perron (PP) tests. The Variance Ratio (VR) test was used to assess autocorrelation, while the Wald-Wolfowitz Runs (WWR) test was utilized to examine random walk behavior. While the Runs test indicated some randomness, all other tests rejected their null hypotheses, indicating that the Nomu market is inefficient according to the weak-form model. This inefficiency suggests that historical trading data may predict future prices, which could discourage foreign capital inflows, thereby limiting Nomu's growth and investment appeal. The significance of these findings for regulators and policymakers in formulating market regulations and policies aimed at enhancing Nomu's efficiency and investor appeal cannot be overstated. This paper, with its robust methodology, is believed to be the first to evaluate the Nomu market's weak-form efficiency since its inception, contributing to a better understanding of market dynamics in emerging stock exchanges.

Reinforcement-based processes actively regulate motor exploration along redundant solution manifolds.

From a baby's babbling to a songbird practising a new tune, exploration is critical to motor learning. A hallmark of exploration is the emergence of random walk behaviour along solution manifolds, where successive motor actions are not independent but rather become serially dependent. Such exploratory random walk behaviour is ubiquitous across species' neural firing, gait patterns and reaching behaviour. The past work has suggested that exploratory random walk behaviour arises from an accumulation of movement variability and a lack of error-based corrections. Here, we test a fundamentally different idea-that reinforcement-based processes regulate random walk behaviour to promote continual motor exploration to maximize success. Across three human reaching experiments, we manipulated the size of both the visually displayed target and an unseen reward zone, as well as the probability of reinforcement feedback. Our empirical and modelling results parsimoniously support the notion that exploratory random walk behaviour emerges by utilizing knowledge of movement variability to update intended reach aim towards recently reinforced motor actions. This mechanism leads to active and continuous exploration of the solution manifold, currently thought by prominent theories to arise passively. The ability to continually explore muscle, joint and task redundant solution manifolds is beneficial while acting in uncertain environments, during motor development or when recovering from a neurological disorder to discover and learn new motor actions.

BIG DATA AND EXCHANGE RATE EFFICIENCY IN NIGERIA: ANY ROLE FOR INVESTMENT SENTIMENTS?

The efficiency of the exchange rate is a strong indicator to determine appropriate exchange rate returns. This study engages the use of big data to investigate exchange rate efficiency in Nigeria and further examine the role of investment sentiments. The study employs both the unit-root and variance ratio tests. Also, the granger causality test was employed to investigate the direction of causality. The data spanned 4,992 daily observations each for exchange rate and investment sentiments and cover the daily period 12/10/2001 – 5/13/2022. For further interrogation, quarterly and yearly data frequencies of these two variables were employed in order to explain the varieties of big data. With recourse to the effect of big data, results show that the exchange rate exhibits a random walk behaviour in Nigeria only for daily and quarterly data frequencies and not for the yearly data frequency. However, the causality test indicates that investors' speculations do not affect exchange rate dynamics in the country. As the exchange rate is found efficient in Nigeria, the monetary authority is enjoined to promote real-time information about the exchange rate to deflect undue speculations by investors. This study implies that the monetary authority in Nigeria should model exchange rate efficiency around its intrinsic data-generating process.

Asymmetric market efficiency of the Eurozone using the MF-DFA: a comparison between global financial crisis and COVID-19 era

PurposeThis research paper aims to explore asymmetric market efficiency of the 13 Euro countries, i.e. Austria, Belgium, Finland, France, Germany, Greece, Ireland, Italy, Netherland, Portugal, Slovakia, Slovenia and Spain, concerning the period before global financial crisis (GFC), after GFC and period of COVID-19 pandemic.Design/methodology/approachMultifractal detrended fluctuation analysis (MF-DFA) is applied to examine the persistence and anti-persistency. It also discusses the random walk behavior hypothesis of these 13 countries non-stationary time series. Additionally, generalized Hurst exponents are applied to estimate the relative efficiency between short- and long-run horizons and small and large fluctuations.FindingsThe current study results suggest that most countries' markets are multifractal and exhibit long-term persistence in the short and long run. Moreover, the results with respect to full sample confirm that Portugal is the most efficient country in short run and Austria is the least efficient country. However, in long run, Austria appeared to be highly efficient, and Slovakia is the least efficient. In the pre-GFC period, Greece is said to be the relatively most efficient market in the short run, whereas Austria is the most efficient market in the long run. In the case of Post-GFC, Netherland and Ireland are the most efficient markets in short and long run, respectively. Lastly, COVID-19 results indicate that Finland's stock market is the most efficient in short run. Whereas, in the long run, the high efficiency is illustrated by Germany. In contrast, the most affected stock market due to COVID-19 is Belgium.Originality/valueThis study will add value to the present knowledge on efficient market hypothesis (EMH) with the MF-DFA approach. Also, with the MF-DFA approach, potential investors will be capable of ranking the stock markets of Eurozone countries based on their efficiency in the period before and after GFC and then specifically in the period of COVID-19.

Evaporation-driven directed motion of droplets on the glass

Manipulating and guiding the droplet movement on a substrate by vapor-mediated actuation is a new approach for microfluidic devices. In this work, the evaporation-driven self-propulsion of the droplets such as N,N-dimethylacetamide, and N-methylformamide on the glass is demonstrated. By the addition of the Silwet surfactant, the aforementioned liquids change from total wetting to partial wetting and the two-component droplet is formed on the glass. The random walk behavior of the droplet can be directed by the linear and octagonal runways on a substrate, of which both hydrophobic sides are constructed by the polysilazane coating. The long-range transport of the droplet is observed, and the effect of the Silwet concentration on the self-propelled velocity is examined. Finally, the evaporation-driven mechanism is investigated by employing the open and closed environments for one-droplet and two-droplet systems. It is found that the self-propelled behavior of Silwet-laden droplets vanishes in the closed environment.

Understanding the Structure and Rheology of Galactomannan Solutions with Coarse-Grained Modeling

Understanding and predicting the behavior of aqueous solutions of galactomannans is of significant interest for a variety of fields and applications, such as hydraulic fracturing in the petroleum industry and thickening in the food industry. Herein, we devise and validate a coarse-grained (CG) model of guar gum─a galactomannan─and study its structure and rheology in aqueous solution. The CG model is computationally efficient, utilizing an implicit solvent and one CG bead per monosaccharide, and exhibits good quantitative agreement with experiments with respect to the radius of gyration of the chain, aggregate structure in solution, and solution viscosity. Using both simulations and experiments, we show that the behavior of guar gum in water can be explained by scaling theory for polymers in good solvents. Namely, the radius of gyration of the chain shows a self-avoiding random walk behavior, and the solution specific viscosity ηsp exhibits the expected power-law scaling with the reduced concentration in different solution regimes (i.e., ηsp ∼ (c[η])1.3 in semidilute and ηsp ∼ (c[η])3.9 in entangled solutions). By extending our CG model to other galactomannans with different mannose-to-galactose ratios (M/G ratios) and different arrangements of unsubstituted mannose units, we show that high M/G ratios and long blocks of unsubstituted mannose units are responsible for aggregate formation in galactomannan solutions, which can reduce solution viscosity moderately.

Trend reversal and alpha generation by hedge funds

ABSTRACT Hedge funds use different strategies to outperform benchmark indices by undertaking complex, versatile and dynamic techniques. These strategies may however create alpha generating opportunities in the time series that may be contrary to the existence of random walk behaviour. Tests using unit-root structural break analysis led to rejection of random walk hypothesis favouring trend stationarity or mean reversion. Additional tests including structural breaks are conducted to establish persistence of alpha generating opportunities for twelve different hedge fund indices. For some strategies, profitable prospects occur only before the break date, as evidence suggests a paradigm shift between the two periods.

Can wavelets produce a clearer picture of weak-form market efficiency in Bitcoin?

This study proposes a new approach for testing for random walk behavior in daily Bitcoin returns (19/07/2010–03/03/2022) by contextualizing the Dickey-Fuller test in time-frequency space using continuous complex wavelet transforms. By splitting our full sample into smaller sub-sample periods segregated by Bitcoin halving dates, we find that Bitcoin returns are most predictable or least market efficient (i) at higher frequency or short-run cycles of between 2 and 16 days, (ii) between November-February months, (iii) during ‘bubble’ periods, (iv) across the consecutive halving dates, (v) during the ‘Black Swan event’ caused by financial market turmoil arising from the COVID-19 pandemic, and (vi) subsequent to the announcements of new COVID-19 variants. Altogether, our findings have important policy implications for different stakeholders in Bitcoin markets.

Frequency Stability Transfer in Passive Mode-Locked Quantum-Dash Laser Diode Using Optical Injection Locking

In this paper, we present an experimental study of the metrological stabilization of a solid-state frequency comb for embedded metrology applications. The comb is a passively mode-locked laser diode based on InGaAs/InP Quantum-dash structure emitting optical lines into a 9 nm bandwidth centered at 1.55 $\mu$m with a repetition rate of 10.09 GHz. The frequency stabilization is achieved by optical injection locking of the comb with an external cavity laser diode referenced onto a metrological frequency standard. One observes the transfer of the spectral purity from the injection laser to the neighbouring modes of the injected one as well as the transfer of stability to the adjacent modes. The measurement of the long term stability highlights a frequency noise with random walk behavior specific of the passive mode locking process. Demonstration of sidebands of the injection laser at the repetition frequency of the comb also makes it possible to propose a transfer mechanism and to consider a complete stabilization of the frequency comb at a metrological stability level.

Forecasting Technique Bridges Gap Between Material Balance and Reservoir Simulation

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 203941, “Bridging the Gap Between Material Balance and Reservoir Simulation for History Matching and Probabilistic Forecasting Using Machine Learning,” by Nigel H. Goodwin, Essence Analytics. The paper has not been peer reviewed. Currently, building and maintaining a reservoir simulation model is the bottleneck for reservoir decision support. In this paper, the authors describe an approach for creating a digital twin that is informed by the data itself based on established physics modeling and uses a fully probabilistic Bayesian approach. Statistical Time-Series Modeling and Physics Models Differences Between Statistical Models and Physics Models. The statistical model includes an error term for each timestep. In practice, this means that the predicted fit to historic data is very good and the error terms are small, but forecasting can have a major funneling effect. Many methods exist for forecasting of statistical models, but the increasing uncertainty of prediction over time, and the large magnitude of uncertainty, is a common feature. The error term implies a future random walk behavior. When forecasting, multiple single-step integrations are being performed in which each step has its own error. Where known values of exogenous variables exist in the future, prediction uncertainty can be greatly ameliorated. In contrast, the differential algebraic equations (DAE) for a physical model have no error term for each timestep. Each time the model is run with the same parameter values, the same results are obtained. The error in reservoir simulation lies in measurement error and model error. Errors exist in individual historical measurements (or maybe systemic errors), and the model and its parameters are known to be wrong. For forecasting, a range of models and model parameter values are used—ideally, based around a robust Monte Carlo approach. An ensemble of forecasts can then be generated that represent uncertainty.

Skew-normal log-volatility model of road surface profile

Road roughness-induced vibrations are the main source of fatigue damage accumulation in vehicles. Hence, road surface profile modeling and simulation are of high importance when it comes to vehicle fatigue damage assessment.This research focuses on uncovering the statistical distributions that describe the characteristics of road loads that affect fatigue damage accumulation. Particle filtering is deployed to estimate the log-volatility of the road surface profile by assuming a random walk behavior for the hidden log-volatility. The skew-normal distribution with three parameters is fitted to the estimated log-volatility. The inferred parameters are used for synthesizing artificial road profiles used in a vehicle simulation to calculate fatigue damage indices (FDI). The new road profile model outperforms all the previous models in predicting the real road FDI.

Efficiency Drifts in Euronext Stock Indexes Returns

This paper intends to assess and test long-term memory in the Euronext stock indexes returns in the search for fractal dynamics that refute the random walk hypothesis. The Hurst exponents estimated through Rescaled-Range and Detrended Fluctuation Analysis evidence long memory in the form of persistence for all markets, with the exception of CAC 40 by the DFA. However, the Rescaled-Range Tests neither reject the absence of long dependency nor reject the existence of short dependency. On the contrary, the Fractional Differencing Test supports the presence of persistence in the PSI 20, ISE 20 and OBX indexes. This suggests that these markets are more prone to predictability, but also trends that may be unexpectedly disrupted by discontinuities, exhibiting dynamics incompatible with random walk behavior and providing evidence against the weak form of efficiency and validity of the asset pricing models.

Locally Scaled and Stochastic Volatility Metropolis– Hastings Algorithms

Markov chain Monte Carlo (MCMC) techniques are usually used to infer model parameters when closed-form inference is not feasible, with one of the simplest MCMC methods being the random walk Metropolis–Hastings (MH) algorithm. The MH algorithm suffers from random walk behaviour, which results in inefficient exploration of the target posterior distribution. This method has been improved upon, with algorithms such as Metropolis Adjusted Langevin Monte Carlo (MALA) and Hamiltonian Monte Carlo being examples of popular modifications to MH. In this work, we revisit the MH algorithm to reduce the autocorrelations in the generated samples without adding significant computational time. We present the: (1) Stochastic Volatility Metropolis–Hastings (SVMH) algorithm, which is based on using a random scaling matrix in the MH algorithm, and (2) Locally Scaled Metropolis–Hastings (LSMH) algorithm, in which the scaled matrix depends on the local geometry of the target distribution. For both these algorithms, the proposal distribution is still Gaussian centred at the current state. The empirical results show that these minor additions to the MH algorithm significantly improve the effective sample rates and predictive performance over the vanilla MH method. The SVMH algorithm produces similar effective sample sizes to the LSMH method, with SVMH outperforming LSMH on an execution time normalised effective sample size basis. The performance of the proposed methods is also compared to the MALA and the current state-of-art method being the No-U-Turn sampler (NUTS). The analysis is performed using a simulation study based on Neal’s funnel and multivariate Gaussian distributions and using real world data modeled using jump diffusion processes and Bayesian logistic regression. Although both MALA and NUTS outperform the proposed algorithms on an effective sample size basis, the SVMH algorithm has similar or better predictive performance when compared to MALA and NUTS across the various targets. In addition, the SVMH algorithm outperforms the other MCMC algorithms on a normalised effective sample size basis on the jump diffusion processes datasets. These results indicate the overall usefulness of the proposed algorithms.