Lead-lag Effect Research Articles

Lead–lag effect of research between conference papers and journal papers in data mining

AbstractThe examination of the lead–lag effect between different publication types, incorporating a temporal dimension, is very significant for assessing research. In this article, we introduce a novel framework to quantify the lead–lag effect between the research topics of conference papers and journal papers. We first identify research topics via the text‐embedding‐based topic modeling technique BERTopic, then extract the research topics of each time slice, construct and visualize the similarity matrix of topics to reveal the time‐lag direction and finally quantify the lead–lag effect by four proposed indicators, as well as by average influence topic similarity comparison maps. We conduct a detailed analysis of 19,166 bibliographic data for top conference papers and journal papers from 2015 to 2019 in the data mining field, calculate the similarity of topics obtained by BERTopic between each time slice divided by quarters. The results show that journal paper topics lag behind conference paper topics in the data mining field. The most significant lead–lag effect is 2.5 years, with approximately 33.45% of topics affected by this lag. The methodology presented here holds potential for broader application in the analysis of lead–lag effects across diverse research areas, offering valuable insights into the state of research development and informing policy decisions.This article is categorized under: Application Areas > Science and Technology

Bitcoin returns and YouTube news: a behavioural time series analysis

ABSTRACT This study investigates whether investor’s sentiment and attention information collected via YouTube can improve bitcoin return forecasts. Accordingly, we collected daily data over the period 2017–2023, covering calm and turbulent periods marked by different types and episodes of emotions. Unlike previous studies, we used YouTube videos to propose two sentiment proxies: investor attention to YouTube (daily number of YouTube video views) and investor sentiment on YouTube (number of positive and negative videos on YouTube). Interestingly, we break down both attention and sentiment per subject. Econometrically, we assess lead-lag effects between sentiment/attention and bitcoin return using causality tests and Vector Auto-regressive (VAR) model. We also evaluate the forecasting power of YouTube attention/sentiment data using a deep learning LSTM model. Our study shows two main results. First, we find lead-lag effects between bitcoin returns and per subject investor’s attention and sentiment proxies. Second, we show that our deep learning LSTM model relying on the information provided by attention and sentiment supplants benchmark Buy and Hold Strategy to forecast future bitcoin returns.

Crude oil futures and the short-term price predictability of petroleum products

This study investigates the lead-lag effect between crude oil futures and petroleum products from the view of price predictability. The findings of the Granger causality test provide evidence for the leading role of crude oil futures. Further empirical results show that crude oil futures demonstrate strong predictive power for petroleum products both in- and out-of-sample, and this predictability is a short-term phenomenon. During economic recessions, petroleum product prices exhibit stronger predictability, with out-of-sample R2 reaching a maximum of 39.161 %. Additionally, the outstanding forecasting performance remains robust across a range of settings. Finally, two potential economic sources underlying the lead-lag relationship between crude oil and petroleum product markets are provided: the differential response speed to common energy market information and the perspective of raw material supply.

Multiproxy studies on the spatially distinct surface samples to reconstruct palaeoecology and palaeoclimate from the Core Monsoon Zone of India

Palaeoclimatic reconstructions necessitate an understanding of the various biotic and abiotic responses to develop a modern analogue. The interpretation and calibration of the modern data allow us to check the lead-lag effect in different proxy parameters, which could be applied for robust palaeoclimatic reconstructions. We, in the present study, analysed pollen and diatom (biotic proxies), as well as grain size, magnetic susceptibility and geochemistry (abiotic components) of the modern soil samples, collected from the states of Madhya Pradesh and Chhattisgarh, the Core of the Monsoon Zone (CMZ) in central India. The weathered materials of the Palaeocene Cretaceous extrusive rocks and sedimentary rocks of the Late Triassic to the Upper Carboniferous are underlying the soil cover in these areas, respectively. The study revealed that the overall pollen and diatom preservation is comparatively good in the areas where the Palaeocene Cretaceous extrusive rocks are found except for the areas of human settlements, whereas the preservation of pollen and diatom was comparatively poor in areas where sedimentary rocks of the Late Triassic to Upper Carboniferous are found. The most plausible reason for this difference is the availability of nutrients which are supplied more abundantly by the easily weatherable Deccan basalt rocks compared to their sedimentary counterparts. The present study will serve as baseline information about biotic-abiotic interactions operating in the central Indian Core Monsoon Zone (CMZ). Since the intensity and duration of the rainfall in the CMZ are largely governed by the annual Indian summer monsoon rainfall (AISMR), therefore, the present study could help trace the weak or intense monsoon periods (break and active spells, respectively) of the past hundreds to thousands of years by studying the sediment profiles/cores.

Tokenomics in the Metaverse: understanding the lead–lag effect among emerging crypto tokens

The convergence of blockchain and immersive technologies has resulted in the popularity of Metaverse platforms and their cryptocurrencies, known as Metaverse tokens. There has been little research into tokenomics in these emerging tokens. Building upon the information dissemination theory, this research examines the role of trading volume in the returns of these tokens. An empirical study was conducted using the trading volumes and returns of 197 Metaverse tokens over 12 months to derive the latent grouping structure with spectral clustering and to determine the relationships between daily returns of different token clusters through augmented vector autoregression. The results show that trading volume is a strong predictor of lead–lag patterns, which supports the speed of adjustment hypothesis. This is the first large-scale study that documented the lead–lag effect among Metaverse tokens. Unlike previous studies that focus on market capitalization, our findings suggest that trade volume contains vital information concerning cross-correlation patterns.

Time trajectories and within-subject correlations of matrix metalloproteinases 3, 8, 9, 10, 12, and 13 serum levels and their ability to predict mortality in polytraumatized patients: a pilot study

BackgroundManaging polytrauma victims poses a significant challenge to clinicians since applying the same therapy to patients with similar injury patterns may result in different outcomes. Using serum biomarkers hopefully allows for treating each multiple injured in the best possible individual way. Since matrix metalloproteinases (MMPs) play pivotal roles in various physiological processes, they might be a reliable tool in polytrauma care.MethodsWe evaluated 24 blunt polytrauma survivors and 12 fatalities (mean age, 44.2 years, mean ISS, 45) who were directly admitted to our Level I trauma center and stayed at the intensive care unit for at least one night. We determined their MMP3, MMP8, MMP9, MMP10, MMP12, and MMP13 serum levels at admission (day 0) and on days 1, 3, 5, 7, and 10.ResultsMedian MMP8, MMP9, and MMP12 levels immediately rose after the polytrauma occurred; however, they significantly decreased from admission to day 1 and significantly increased from day 1 to day 10, showing similar time trajectories and (very) strong correlations between each two of the three enzyme levels assessed at the same measurement point. For a two-day lag, autocorrelations were significant for MMP8 (− 0.512) and MMP9 (− 0.302) and for cross-correlations between MMP8 and MMP9 (− 0.439), MMP8 and MMP12 (− 0.416), and MMP9 and MMP12 (− 0.307). Moreover, median MMP3, MMP10, and MMP13 levels significantly increased from admission to day 3 and significantly decreased from day 3 to day 10, showing similar time trajectories and an (almost) strong association between every 2 levels until day 7. Significant cross-correlations were detected between MMP3 and MMP10 (0.414) and MMP13 and MMP10 (0.362). Finally, the MMP10 day 0 level was identified as a predictor for in-hospital mortality. Any increase of the MMP10 level by 200 pg/mL decreased the odds of dying by 28.5%.ConclusionsThe time trajectories of the highly varying individual MMP levels elucidate the involvement of these enzymes in the endogenous defense response following polytrauma. Similar time courses of MMP levels might indicate similar injury causes, whereas lead–lag effects reveal causative relations between several enzyme pairs. Finally, MMP10 abundantly released into circulation after polytrauma might have a protective effect against dying.

Considering momentum spillover effects via graph neural network in option pricing

AbstractTraditional options pricing relies on underlying asset volatility and contract properties. However, asset volatility is affected by the “lead–lag effects,” known as the “momentum spillover effect.” To address this, we propose a proxy measuring correlated options' influence based on maturity date. Findings indicate that 1‐day‐lagged proxy indicators positively impact option returns. Furthermore, to capture the dynamic effects of correlated options, we introduce a deep graph neural network‐based model (GNN‐MS). Empirical results on Shanghai Stock Exchange 50 exchange‐traded fund options reveal GNN‐MS significantly outperforms classics, enhancing root‐mean‐square error by at least 8.81%. This study provides novel insights into option pricing considering momentum spillover effects.

Equal Employment Opportunity in Supply Chains

An equal employment opportunity (EEO) policy is an essential practice of workplace diversity, equity, and inclusion. Using a novel measure of workplace EEO based on the textual analysis of online job postings, we show a lead–lag pattern of EEO practice between principal customers and their dependent suppliers. The leadlag effect is more pronounced when principal customers enjoy greater bargaining power, consistent with the view that dependent suppliers adjust EEO policies to cater to customer preferences. We use the 2017 Diversity Campaign of the “Big Three” institutional investors as our main identification strategy to support a causal interpretation. We demonstrate that the diffusion of EEO policies along supply chains improves the likelihood of retaining female employees, implying the presence of social benefits in addition to economic advantages.

The relationship between Chinese and FOB prices of rare earth elements – Evidence in the time and frequency domain

This paper investigates the relationship between domestic and export prices of rare earth elements (REEs) taking into account both, the time and the frequency domain using Granger causality tests and cross wavelet analysis. We find areas of significant comovement between both, returns and volatility of domestic and export prices, with the degree of connection being low at higher frequencies, increasing over medium frequencies and lower frequencies. Moreover, the lead-lag effects between the Chinese and the export market are time-varying and heterogenous, with no particular market as the leader. The analysis of volatility transmission shows that own-market volatility rather than spillover from the Chinese to the export market or vice versa accounts for the biggest share of volatility. While volatility transmission shows substantial variation over time, the Chinese market is generally a net giver of volatility to the export market, especially at the medium frequency band. Accordingly, market participants buying REEs in the export market might want to track Chinese prices as leading indicators of price fluctutations due to spillovers.

Production complementarity and information transmission across industries

Economic theory suggests that production complementarity is an important driver of sectoral co-movements and business cycle fluctuations. We operationalize this concept using a measure of production complementarity proximity (COMPL) between any two companies. We show firms from different industries but are closely aligned in COMPL exhibit strong co-movement in their operating, investing, and financing activities, as well as quarterly earnings revisions and monthly returns. We further document a lead-lag effect in their returns, such that a long-short strategy based on recent COMPL peer returns yields a monthly 6-factor alpha of 122 basis points. This inter-industry momentum spillover effect is not explained by other network-based mechanisms, such as shared analyst coverage. We conclude information transmission takes place along complementarity networks, but stock prices do not update instantaneously.

The profitability of lead–lag arbitrage at high frequency

Any lead–lag effect in an asset pair implies that future returns on the lagging asset have the potential to be predicted from past and present prices of the leader, thus creating statistical arbitrage opportunities. We utilize robust lead–lag indicators to uncover the origin of price discovery, and we propose an econometric model exploiting that effect with level 1 data of limit order books (LOBs). We also develop a high-frequency trading strategy based on the model predictions to capture arbitrage opportunities. The framework is then evaluated on six months of DAX 30 cross-listed stocks’ LOB data obtained from three European exchanges in 2013: Xetra, Chi-X, and BATS. We show that a high-frequency trader can profit from lead–lag relationships because of predictability, even when trading costs, latency, and execution-related risks are considered.

A seesaw effect in the cryptocurrency market: Understanding the return cross predictability of cryptocurrencies

This paper investigates the intraday return cross-predictability of cryptocurrencies. In contrast to the positive lead–lag effect for stocks, we document a negative lead–lag effect in the cryptocurrency market. Specifically, the large coins negatively predict the other coins but the small coins rarely predict the large coins. A trading strategy that exploits the cross-predictability via the Least Absolute Shrinkage and Selection Operator (LASSO) yields highly significant profits across major cryptocurrency exchanges even in the presence of realistic transaction costs.

Significant wave height prediction based on dynamic graph neural network with fusion of ocean characteristics

Significant wave height (SWH) is one of the core parameters for wave and accurate prediction of SWH is of great importance for ocean resource assessment. In this paper, we propose a new multi-characteristic and multi-node SWH prediction model(MCMN). The model considers the lead–lag effect among ocean characteristics and utilizes time lag correlation to automatically learn advanced indication information. For the temporal features, temporal correlations are extracted from high-dimensional spatial features efficiently in parallel using Temporal Convolutional Network(TCN). Additionally, the dependencies between nodes are modeled as the joint result of stable long-term patterns and dynamic short-term patterns. To obtain these dependencies, we introduce a novel dynamic graph neural network. Compared to previous SWH predictions focused solely on individual nodes, this model allows us to more fully explore the spatio-temporal dependencies between the nodes by capturing both long-term and short-term spatio-temporal relationship patterns among the nodes. Experiments were conducted with 120 nodes in the South China Sea and East China Sea, respectively. The results show that the model provides reliable predictions. Finally, we compare with five deep learning models, and the results show that our model has better performance in multi-node and multi-step SWH prediction.

Evolution of short-term contrarian profits

PurposeThe purpose of this paper is to study the US stock market and try to explain why short-term contrarian profits have largely disappeared in the past two decades.Design/methodology/approachIn this work, the authors decompose the short-term contrarian profits into cross-sectional variations, firm-level overreactions and lead-lag effects to study the changes in their shares. Then, the authors study the behavior of the subgroups in the winner and loser subportfolios of contrarian investment strategies.FindingsThe authors find that short-term contrarian profits have largely vanished since 2000. Changes in the shares of the three components of contrarian profits, which are cross-sectional variations, firm-level overreactions and lead-lag effects, are not the main reason for the disappearance of contrarian profits in the past two decades. Instead, the disappearance of short-term contrarian profits is primarily due to the heterogeneous evolution of subgroups in the portfolio, which leads to a decrease in the overall level of overreactions that drive the contrarian profit.Originality/valueThe work explains the disappearance of short-term contrarian profits in the US stock market.

The predictability of iron ore futures prices: A product‐material lead–lag effect

AbstractThis study investigates the lead–lag effects between product futures and raw material futures. Results show that returns on product futures lead returns on raw material futures: lagged product futures returns can significantly predict raw material futures returns in‐ and out‐of‐sample. This product‐material lead–lag effect is mainly driven by bad news and is a short‐term phenomenon. Moreover, returns on product futures, especially those based on bad news, can provide substantial economic gains to investments in raw material futures. The lead–lag effect is associated with frictions in the information diffusion from product futures markets to raw material futures markets and the slower response of raw material futures markets to common market information.

Disseminating information across connected firms — Analyst site visits can help

When information is disseminated across fundamentally connected firms at a fast pace, limited investor attention should be mitigated. We investigate whether analyst site visits to public firms have any impact on the lead–lag effect, a well-known phenomenon induced by limited investor attention. We find that analyst site visits to a focal firm’s connected firms almost eliminate the lead–lag effect between the visited connected firms and the focal firm at weekly frequency. The Fama–French five-factor alpha drops from 0.365% (t-statistic=3.26) per week for the strategy based on lagged returns of a firm’s non-visited connected firms to -0.082% (t-statistic=-0.97) per week for the strategy based on lagged returns of visited connected firms. We further provide evidence that analyst site visits to connected firms help improve forecast accuracy for the forecasted firm. These results indicate that analyst site visits help disseminate information across connected firms.

High frequency volatility spillover between oil and non-energy commodities during crisis and tranquil periods.

In this article, we scrutinize volatility spillover between oil and individual non-energy commodities during crisis and non-crisis periods. We use high-frequency data to capture the effects of both the global financial crisis (2008) and the COVID-19 pandemic between 2008 and 2022. To this end, we utilize wavelet coherence analysis to diagnose the magnitudes of dynamic co-movements and lead-lag effects between commodities. Our results provide evidence of strong coherence between oil and the majority of individual non-energy commodities during both crises. Precious metals were generally found to exhibit heightened levels of co-movement with oil as opposed to other non-energy commodities. On the other hand, weak co-movements were found between oil and a few commodities, namely soy, wheat, zinc, and tin. The lead-lag effects of oil on agricultural commodities, base metals, and precious metals were evident, especially during crisis periods. However, aluminium and precious metals, especially gold, silver, and palladium, also had a lead-lag effect on oil at different points in time, including during the pandemic. We further utilize dynamic frequency-domain connectedness for capturing pairwise volatility spillover indices, with the results providing evidence of heightened volatility spillovers during turbulent times. Our findings have significant implications for retail investors, portfolio managers, and policymakers.

Money Growth and Inflation: A Wavelet Analysis for a Monetization Approach to COVID 19

COVID 19 pandemic will cause a global economic contraction. The magnitude of shocks will impact the fiscal deficit and the public debt in all Countries. This paper provides an insight into the dynamic relationship between money growth and inflation for several Countries. The Morlet wavelet coherence model is employed since it allows the simultaneous examination of lead-lag effects and co-movements between the couple of variables in the different Countries. The overall results do not evidence a clear impact of the money growth on inflation. This outcome has relevant economic policy consequences considering that the monetization of the public debt can sustain the economic recovery. Thus, the pure monetary nature of the inflationary phenomenon can hardly be considered as a constraint in designing the measures to counteract the issue.

Effects of life events and attitudes on vehicle transactions: A dynamic Bayesian network approach

Individual and household life events are interdependent and influence mobility-related decisions at different levels over time. This paper developed an integrated dynamic model to capture the interdependences among life events, with a special focus on vehicle transactions. Particular attention was paid to the inclusion of vehicles’ characteristics such as the age, fuel type, and size of cars, which are pertinent to emission forecast. A dynamic Bayesian network (DBN), containing individual and household characteristics and latent attitudes toward car ownership and use alongside life events, was employed to study the interdependences. The temporal relationships among life events and lead-lag effects were also captured in the DBN. The longitudinal survey data “the Netherlands Mobility Panel (MPN)” from 2013 to 2018 was used to train and test the DBN. The analysis results confirm the dynamic interdependences between vehicle transactions and other life events and reveal noticeable associations between attitudes and purchase decisions. It is found that several life events (e.g., “Birth of a baby”, “Marital status change”) have concurrent or varied lag-effects on vehicle transaction decisions. The validation indicates that the proposed DBN approach has a high predictive accuracy of vehicle transaction decisions and other life events.

Can Collective Emotions Improve Bitcoin Volatility Forecasts?

This paper extends the study of Bourghelle et al. (2022) to check whether collective emotions could help forecast bitcoin volatility over the period 2018-2021. To this end, we first assess whether consideration of investor sentiment and collective emotions can give us clearer insights into bitcoin dynamics over the period in question and whether it can help to explain the different shifts in price. Formally, we ran causality tests and, as in Bourghelle et al. (2022), built a two equation nonlinear vector autoregressive (VAR) model to assess for further lead-lag effects between bitcoin volatility and collective emotions. Second, we proposed in-sample forecasts of bitcoin volatility to test whether it would be possible to improve our forecasts by taking investors’ emotions and sentiment into account. Our findings show that market sentiment and investors’ emotions provide useful information that can explain shifts, structural breaks, and changes in bitcoin volatility. Further, collective emotions improve bitcoin volatility forecasting as our nonlinear model, including emotions-related news, supplants the benchmark linear model.