Detrended Cross-correlation Analysis Research Articles

Analysis on Time-Series Data from Movie Using MF-DCCA Method and Recurrent Neural Network Model Under the Internet of Things.

The present work aims to analyze the time-series data (TSD) from movies and support constructing the movie recommendation system. Referencing the Internet of Things (IoT) technology as the framework, a time-series data analysis system for movies is built based on the recurrent neural network (RNN) and multifractal detrended mobility cross-correlation analysis (MF-DCCA) method. First, the traditional RNN model is improved by replacing the conventional convolution operation with spatial adaptive convolution. Specifically, an additional convolution layer is used to obtain the position parameters required for adaptive convolution to improve the model performance to capture the characteristics of spatial-temporal transformation. Then, the MF-DCCA method is optimized to reduce the interference of noise signals to the analysis processing of TSD from movies. Finally, the TSD analysis system is tested for performance verification. The test results indicate that the method proposed here has outstanding stability and runs smoothly. When the prediction scheme is long short-term memory (LSTM) (L = 20), the similarity of the LSTM (L = 20) network under one frame is 0.977; the similarity of the LSTM (L = 20) network under nine frames is 0.727. This system provides a specific idea for applying the RNN model and MF-DCCA method in analyzing TSD from movies.

The dynamic relationship of cryptocurrencies with supply chain and logistics stocks – the impact of COVID-19

PurposeThe COVID-19 pandemic is known to have affected the logistics and supply chains; however, there is no adequate empirical evidence to prove in which way it has affected the relationship between the stocks related to this field with the corresponding cryptocurrencies. This paper aims to test the dynamic relationship of cryptocurrencies with supply chain and logistics stocks.Design/methodology/approachIn this paper, the author tests the causal and long-run relationship between logistics and supply chain stocks with the corresponding cryptocurrencies related to these fields, or those that are known to exhibit characteristics that can be utilized by these fields, testing also whether the COVID-19 pandemic affected this relationship. To do so, the author performs the variable-lag causality to test the causal relationship, and examines if this relationship changed due to COVID-19. The author then implements the multifractal detrended cross-correlation analysis to investigate the characteristics of a possible long-run relationship, testing also whether they changed due to COVID-19.FindingsThe results indicate that there is a positive long-run relationship between each logistics and supply chain stocks and the corresponding cryptocurrencies, before and also during COVID-19, but during COVID-19 this relationship becomes weaker, in most cases. Moreover, before COVID-19, the majority of the cases indicate a causal direction from cryptocurrencies to the stocks, while during COVID-19, the causal relationships decrease in multitude, and most cases unveil a causal direction from the stocks to cryptocurrencies.Originality/valueThe causal pattern changed during COVID-19, and the long-run relationship became weaker, showing a change in the dynamics in the relationship between logistics and supply chain stocks with cryptocurrencies.

The Global Transmission of Stock Market: A Spatial Analysis

The stock markets, exhibiting complex self-correlation or cross-correlation over a broad range of time scales, are correlated not only in time but also in space. The conventional spatial weight matrix in the econometric analysis is short of economic relation between nonadjacent economic entities. Therefore, this paper applies the detrended cross-correlation analysis coefficient and partial correlation coefficient to analyze the global spatial interaction. This study computes the spatial Moran’s I value by the two types of weight matrix for the 15 typical stock indices around the world, to explore the spatial agglomeration phenomenon. Then, the Spatial Durbin Model is applied to investigate the transmission of the stock market. The result from the Moran’s I value indicates that the 15 typical stock indices are spatially correlated. The result of the Spatial Durbin Model gives the relationship among the closing price, the opening price, the highest price, and the lowest price.

Relative Prices of Ethanol-Gasoline in the Major Brazilian Capitals: An Analysis to Support Public Policies

The use of biomass as an energy source has advanced in recent decades, given the scientific evidence that it is a solution to the environmental problems faced globally. In this context, biofuels derived from biomass have a prominent role. Among the countries where this alternative is the most promising, Brazil stands out, just behind the USA. It is, therefore, necessary to assess whether such a replacement is economically viable. For such an assessment, the behavior of the relative price of bioethanol/gasoline is crucial. In the present work, the degree of temporal persistence of relative prices, considering the existence of shocks to which they are exposed, is evaluated, considering 15 important Brazilian capitals, via the detrended fluctuation analysis (DFA). The degree of correlation is also evaluated through the detrended cross-correlation analysis (DCCA) between fuel prices in São Paulo, the capital of the most populous state and main producer of bioethanol, with the capitals of the 14 states selected for the analysis. The period of analysis takes place between 2004 and 2020. The use of DCCA with sliding windows was recently proposed and we also evaluate DFA dynamically in this way, and this, together with an extended sample in the context of Brazilian fuel prices, represents the main innovations of the present work. We found that the degree of persistence varies significantly depending on the capitals analyzed, which means that price variations are localized and demand regional stimulus policies. Furthermore, it was found that the correlation with São Paulo is less intense in the most geographically distant capitals. Such evidence is important and complementary to infer how integrated the national bioethanol market is, in order to support public policies aimed at its consolidation.

Interplay between stock trading volume, policy, and investor sentiment: A multifractal approach

The stock market is a typical complex system which may sensitively react to a wide range of external factors, such as investor sentiment and government policies. While such influence on stock market bares significance for understanding the financial dynamics, the interplay between policy, investor sentiment, and stock trading volume is still unclear. Focusing on three individual stocks during the US–China trade war, this paper explores the interplay between three time series, namely the policy intensity, investor sentiment from a stock-related social networking site, and the trading volume of each stock. Considering the multifractal nature of these time series, the multifractal detrended cross-correlation analysis is applied and the results suggest strong correlations between any pair of the time series. As such, the policy and investor sentiment are found with impact on the stock trading volume while the investor sentiment is also influenced by the policy. A further mediation analysis reveals that up to 36% of the influence from the policy to the stock trading volume is mediated by the investor sentiment. Accordingly, the study uncovers and quantifies the path of the policy’s influence on stock trading volume, thereby provides insights for understanding the dynamics of the complex financial systems.

Cross-correlation and forecast impact of public attention on USD/CNY exchange rate: Evidence from Baidu Index

We applied the multifractal detrended cross-correlation analysis (MF-DCCA) method to investigate the cross-correlation behaviour between the USD/CNY exchange rate (UCER) and the Baidu index (Chines netizens’ behaviour big data, BI). We observed a significant anti-persistent cross-correlation between UCER and BI through MF-DCCA testing. Based on the cross-correlation between UCER and BI, a hybrid deep learning model WOA-STL-BI-LSTM combining network big data, decomposition and integration techniques, optimisation algorithms and LSTM is constructed to predict UCER returns, where BI is used as a potential predictor to predict UCER, and the predictability of BI on UCER was discussed. The experimental results show that BI can improve the forecasting accuracy of UCER returns, but not significantly. WOA-STL-BI-LSTM has a significant improvement in forecasting performance compared with the benchmark model and can be used as an advanced model for UCER returns forecasting. This paper provides a new approach and evidence for the study of UCER.

Multifractal Analysis of the Impact of COVID-19 on NASDAQ, CIOPI, and WTI Crude Oil Market

In this paper, we explore the impact of COVID-19 on auto-correlations and cross-correlations among NASDAQ stock index of the USA, China iron ore price index (CIOPI), and West Texas Intermediate Crude Oil price (WTI). To find out the effect of COVID-19 on financial markets, we divide the investigated data series into two sub-periods, i.e., pre-COVID19 period and post-COVID19 period. First, multifractal detrended fluctuation analysis (MF-DFA) of those series shows a general trend of strong multifractality after COVID-19, indicating lower market efficiency after the pandemic shock. Second, multifractal detrended cross-correlation analysis (MF-DCCA) method is employed to examine cross-correlations among NASDAQ, CIOPI, and WTI. The three cross-correlations all increase in post-COVID19. The correlation between NASDAQ and CIOPI increases the most, becoming the strongest correlation among the three cross-correlations in post-COVID19. The surrogate procedure shows that the post-COVID19 cross-correlation multifractalities are mostly due to fat-tail distribution. Third, we use multi-scale multifractal analysis (MMA) to visualize the dynamic behaviors of correlations among the series. The Hurst surfaces of the three cross-correlations have more fluctuation, both at small and large scale in post-COVID19 than that of pre-COVID19. Particularly, the Hurst surface of cross-correlation between NASDAQ and CIOPI exhibits stronger multifractality during the outbreak of COVID-19 than that in both pre-COVID19 and post-COVID19. The above investigations provide helpful insights of relevant market trends.

Application of Multifractal Analysis in Estimating the Reaction of Energy Markets to Geopolitical Acts and Threats

Since the industrial revolution, the geopolitics of energy has been a driver of global prosperity and security, and determines the survival of life on our planet. This study examines the nonlinear structure and multifractal behavior of the cross-correlation between geopolitical risk and energy markets (West Texas Intermediate (WTI), Brent, natural gas and heating oil), using the multifractal detrended cross-correlation analysis. Furthermore, an in-depth analysis reveals different associations of the indices of overall geopolitical risk, geopolitical acts, and geopolitical threats against the four energy products. Based on daily data ranging from 1 January 1985 to 30 August 2021, the findings confirm the presence of nonlinear dependencies, suggesting that geopolitical risk and energy markets are interlinked. Furthermore, significant multifractal characteristics are found and the degree of multifractality is stronger between the overall geopolitical risk and WTI while the lowest degree of multifractality is with Brent. Overall, for the WTI and heating-oil markets, the influence of geopolitical threats is more pronounced rather than their fulfilment. Contrarily, the Brent and natural gas are more correlated to geopolitical acts. Energy products exhibit heterogeneous persistence levels of cross-correlation with all the indicators of geopolitical risk, being more persistent in the case of small fluctuations compared to large fluctuations.

Comparison of Price-Volume Correlation for Some Cryptocurrencies Based on MF-ADCCA

In this paper, we explored the price-volume cross-correlation asymmetric multifractal properties of some cryptocurrencies using multifractal asymmetric detrended cross-correlation analysis (MF-ADCCA). Bitcoin, Ethereum, Tether, Binance Coin, Cardano, and Dogecoin are examples of cryptocurrencies. The empirical results reveal that the price-volume cross-correlations of all six cryptocurrencies display anti-persistent, multifractal, and asymmetric features, although the degree of these characteristics differs depending on the cryptocurrencies and market trends. Specifically, the anti-persistence multifractal of the price-volume cross-correlation is significantly stronger in the upward market than in the downward market for Bitcoin and Ethereum; for the remaining four cryptocurrencies, the opposite result is obtained. Furthermore, Dogecoin price-volume cross-correlation exhibits the most marked anti-persistence and the most significant asymmetric multifractality among the six cryptocurrencies analyzed.

Asymmetric Multifractal Cross-Correlations Between Economic Policy Uncertainty and Agricultural Futures Prices

This paper investigates the fluctuation characteristics and asymmetry of cross-correlations between economic policy uncertainty (EPU) and agricultural futures prices in China and the US by using the multifractal detrended cross-correlation analysis (MF-X-DFA) and multifractal asymmetric detrended cross-correlation analysis (MF-ADCCA). We find that the multifractal cross-correlations exist between EPU and agricultural futures prices, and the cross-correlations are anti-persistent and asymmetric. The anti-persistent cross-correlations in China are all stronger than those in the US. The multifractal degree of cross-correlation between EPU and soybean futures price is lower in China than in the US, while the multifractal degree of cross-correlation between EPU and soybean meal, soybean oil or corn futures price is higher in China than in the US. Moreover, China’s soybean futures price is more susceptible to the upward and downward trends in China’s EPU, while the US soybean meal, soybean oil, and corn futures prices are more susceptible to them in the US EPU.

Robustness of Detrended Cross-correlation Analysis Method Under Outliers Observations

The computation of the bivariate Hurst exponent constitutes an important technique to test the power-law cross-correlation of time series. For this objective, the detrended cross-correlation analysis (DCCA) method represents the most used one. In this paper, we prove the robustness of the DCCA method, where the trend is estimated using the polynomial fitting, to estimate the bivariate Hurst exponent when time series are corrupted by outliers observations. On the other hand, we give the exact polynomial order and a regression region for computing a detrended cross-correlation function to obtain a least square estimator of bivariate Hurst exponent. Our theoretical results are shown by a simulation study on a two-fractional Gaussian noise process corrupted by outliers observations. Additionally, our results are applied to financial time series. The empirical findings results are accompanied by interpretations.

Multifractal risk measures by Macroeconophysics perspective: The case of Brazilian inflation dynamics

This paper examines the Brazilian inflation indexes dynamics using the multifractal detrended fluctuations analysis (MF-DFA) and the multifractal detrended cross-correlation analysis (MF-DCCA). We find that the Brazilian inflation indexes (α0 > 0.5) and the pairs of Brazilian Inflation indexes (Δα > 0.5) display a persistent multifractal behaviour, high complexity and skew symmetries. Also, we propose a novel multifractal risk measure (MR) considering the multifractal cross-correlation measure (MRCC). The higher MR and MRCC values indicate the more complex and persistent analyzed phenomenon. In contrast, the lowest MR value indicates less complexity and less persistence. From a Macroeconophysics perspective, our findings clarify that the dynamics of Brazilian inflation indexes and the pairs of Brazilian inflation indexes genuinely have a robust inertial component that makes inflation last for a long time.

Market-based versus bank-based financial structure in China: From the perspective of financial risk

The evolution of financial structure is not only a persistent issue but also a key factor affecting the financial risk. Concerns over the financial structure evolution and risk prevention have been voiced, but studies documented the interaction between financial risk and financial structure are still in their infancy. Therefore, this study investigates it by using detrended cross-correlation analysis, its multifractal asymmetric version, and detrended cross-correlation based on time-delay. Empirical research finds that the market-based financial structure increases financial risk in China. The characteristics of financial risk and financial structure are asymmetry, multifractal, bidirectional transmission, and nonlinear causality. The financial structure's influence on financial risk is strengthened within 15 months, while financial risk's influence on the financial structure strengths within 7 months. These findings support financial supervisory authorities to seek effective strategies to facilitate the evolution of the financial structure and establish a more stable financial system.

Real-Time Algorithm for Detrended Cross-Correlation Analysis of Long-Range Coupled Processes.

Assessing power-law cross-correlations between a pair – or among a set – of processes is of great significance in diverse fields of analyses ranging from neuroscience to financial markets. In most cases such analyses are computationally expensive and thus carried out offline once the entire signal is obtained. However, many applications – such as mental state monitoring or financial forecasting – call for fast algorithms capable of estimating scale-free coupling in real time. Detrended cross-correlation analysis (DCCA), a generalization of the detrended fluctuation analysis (DFA) to the bivariate domain, has been introduced as a method designed to quantify power-law cross-correlations between a pair of non-stationary signals. Later, in analogy with the Pearson cross-correlation coefficient, DCCA was adapted to the detrended cross-correlation coefficient (DCCC), however as of now no online algorithms were provided for either of these analysis techniques. Here we introduce a new formula for obtaining the scaling functions in real time for DCCA. Moreover, the formula can be generalized via matrix notation to obtain the scaling relationship between not only a pair of signals, but also all possible pairs among a set of signals at the same time. This includes parallel estimation of the DFA scaling function of each individual process as well, thus allowing also for real-time acquisition of DCCC. The proposed algorithm matches its offline variants in precision, while being substantially more efficient in terms of execution time. We demonstrate that the method can be utilized for mental state monitoring on multi-channel electroencephalographic recordings obtained in eyes-closed and eyes-open resting conditions.

Extended detrended cross-correlation analysis of nonstationary processes

We propose an extension of the detrended cross-correlation analysis (DCCA) for signals with highly inhomogeneous structure. The proposed approach evaluates two scaling exponents, one of which characterizes the detrended covariance, and the second exponent quantifies the effects of nonstationarity caused by the distribution of local fluctuations of signal profiles from the trend. Using simulated datasets produced by a system of coupled Lorenz models, we describe entrainment phenomena associated with chaotic synchronization and the role of nonstationarity in their description. The processing of experimental data related to cerebral blood flow in neighboring vessels of different size confirm the general conclusion of this study and shows the possible advantages of the proposed extension of the DCCA-method for diagnosing changes in cooperative dynamics in physiological systems, where the combined effects of dynamics, nonstationarity and noise may occur.

The Evaluation of Mean-Detrended Cross-Correlation Analysis Portfolio Strategy for Multiple risk Assets.

The fractal characteristics of the security market were considered in portfolio strategy optimization. First, the detrended cross-correlation analysis was adopted to measure the fractal correlation of different securities. Second, the fractal correlation was embedded into the mean-variance criterion of the modern portfolio theory. Third, the mean-detrended cross-correlation analysis portfolio strategy of multiple risk assets was constructed, and the analytic solution of the strategy was given. Finally, the evaluation results revealed that the constructed the mean-detrended cross-correlation analysis portfolio strategy clearly improved investment performance, thus achieving the goal of optimizing the multiple risk asset portfolio strategy.

The Cross-Correlation Relationship Between the Real Estate Industry and High-Quality Economic Development

Abstract Based on China’s real estate investment (REI), commercial housing sales (CHS), and GDP data from 2001 to 2020, we studied the dependence, asymmetry, and transmission direction of China’s real estate and economy. Their cross-correlation is demonstrated through the deep cross-correlation analysis (DCCA) algorithm, multifractal asymmetric detrended cross-correlation analysis (MF-ADCCA) algorithm and DCCA algorithm based on a time delay. Both REI and CHS have persistence with GDP. The real estate industry and economy have asymmetric multifractal characteristics. Meanwhile, real estate and the economy have different trends, and they can affect each other with the same length of time delay. This effect diminishes in a short period of time, and there are unsteady fluctuations in cross-correlation persistence in the long term.

Detrended multiple moving average cross-correlation analysis and its application in the correlation measurement of stock market in Shanghai, Shenzhen, and Hong Kong

Based on the traditional detrended moving average cross-correlation analysis (DMCA) method and the detrended multiple cross-correlation analysis (DMC) method, we propose the detrended multiple moving average cross-correlation analysis method (DMMC) to quantify the correlation between non-stationary signals on different time scales. It will be the generalization of the correlation between pairwise time series and can be used to measure the multivariate correlation between the dependent variable y with the others i independent variable x. To validate the DMMC method, we determined the DMMC coefficient into numerical simulations on the standard Gaussian distribution, Lognormal distribution t distribution and ARFIMA long memory random sequences. Then we apply the proposed method to explore the multiple correlations between China’s Shanghai, Shenzhen, and Hong Kong stock markets and analyze the dynamic time-varying characteristics of the correlation with the sliding window approach. The results show that DMMC is suitable to measure multiple cross-correlations. This new method can be further applied to multiple cross-correlation in different areas of knowledge, such as energy, meteorology, medicine, transportation, etc.

Detrended cross-correlation analysis in quantiles between oil price and the US stock market

Previous studies using the detrended cross-correlation analysis (DCCA) did not account for the conditional distribution of time series. To address this shortcoming, this paper extends the DCCA in a quantile-based framework to investigate the relationship between WTI crude oil spot price and the S&P500 index from January 1986 to August 2021 under different oil and stock markets conditions. The approach introduced in this research brings out new and exciting findings on the well-debated oil price-stock markets nexus. By considering four possible scenarios regarding the oil and stock markets conditions, results suggest that the sign and strength of cross-correlations between oil and stock markets vary according to the time scale and market condition. These findings are robust when using daily data and the Brent crude oil spot price. Furthermore, when analyzing the effect of the COVID-19 pandemic, findings generally reveal a rise in correlations between the two markets during the pandemic, suggesting the existence of contagion effects between them. Policy implications for both investors and market regulators are subsequently proposed.

The Discovery of New Drug-Target Interactions for Breast Cancer Treatment.

Drug–target interaction (DTIs) prediction plays a vital role in probing new targets for breast cancer research. Considering the multifaceted challenges associated with experimental methods identifying DTIs, the in silico prediction of such interactions merits exploration. In this study, we develop a feature-based method to infer unknown DTIs, called PsePDC-DTIs, which fuses information regarding protein sequences extracted by pseudo-position specific scoring matrix (PsePSSM), detrended cross-correlation analysis coefficient (DCCA coefficient), and an FP2 format molecular fingerprint descriptor of drug compounds. In addition, the synthetic minority oversampling technique (SMOTE) is employed for dealing with the imbalanced data after Lasso dimensionality reduction. Then, the processed feature vectors are put into a random forest classifier to perform DTIs predictions on four gold standard datasets, including nuclear receptors (NR), G-protein-coupled receptors (GPCR), ion channels (IC), and enzymes (E). Furthermore, we explore new targets for breast cancer treatment using its risk genes identified from large-scale genome-wide genetic studies using PsePDC-DTIs. Through five-fold cross-validation, the average values of accuracy in NR, GPCR, IC, and E datasets are 95.28%, 96.19%, 96.74%, and 98.22%, respectively. The PsePDC-DTIs model provides us with 10 potential DTIs for breast cancer treatment, among which erlotinib (DB00530) and FGFR2 (hsa2263), caffeine (DB00201) and KCNN4 (hsa3783), as well as afatinib (DB08916) and FGFR2 (hsa2263) are found with direct or inferred evidence. The PsePDC-DTIs model has achieved good prediction results, establishing the validity and superiority of the proposed method.