Lead-lag Effect Research Articles

Can Collective Emotions Improve Bitcoin Volatility Forecasts?

This paper extends the study of Bourghelle et al. (2022) to check whether collective emotions could help to 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 they can help to explain the different price fluctuations. 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 our forecasts could be improved by taking investors’emotions and sentiment into account. Our findings show that market sentiment and investors’ emotions provide useful information that can help to explainfluctuations, 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.

HOW GOAL REPRESENTATIONS MODULATE ASSOCIATIONS BETWEEN EVERYDAY AFFECT AND GOAL PURSUIT AMONG OLDER ADULTS

Abstract Individuals differ in the extent to which they represent their goals as hoped-for versus feared states. We examined the role of such goal representations for how everyday affective experiences and goal pursuit are intertwined. When goals are represented as hoped-for states, we expected stronger associations between daily positive affect and goal pursuit. In contrast, when goals are represented as feared states, we expected stronger associations between daily negative affect (particularly fear) and goal pursuit. We used seven days of repeated daily life assessments from 238 older individuals (Age: M = 70.5 years, SD = 5.99, 59-87 years; N = 119 couples). At baseline, participants reported three goals they planned to pursue over the study period and the extent to which each goal referred to something they hoped-for or feared. During the daily life assessments, participants reported their current affective states and momentary goal pursuit (goal engagement and goal progress) three times per day (11 AM, 4 PM, 9 PM). Multilevel analyses regarding participants’ most salient goal provide initial evidence supporting the expected interactions of goal representations on everyday affect–goal pursuit links. Specifically, individuals with a strong hope-focus in their goals engaged in more goal pursuit when positive affect was up than individuals whose goals were low in hope-focus. In contrast, those with feared goals engaged in more goal pursuit when reporting increased fear. Findings are discussed in the context of the possible selves literature. Future analyses will examine lead-lag effects to address the temporal order underlying affect-goal pursuit associations.

Does a firm’s geographic feature matter for stock returns? Evidence from the Chinese A-share market

ABSTRACT This study reveals a new stock return predictability that relates to firms’ geographic features in the Chinese A-share market. Using a text-based measure of the degree of localness to capture the economic ties between firms and their provinces, we find that low-localized firms are slow to incorporate local information into stock prices. Specifically, there is a significant lead-lag effect in stock returns between high- and low-localized firms in the same region, and a portfolio that exploits this pattern can generate a monthly alpha of about 1%. This effect cannot be explained by geographic or industry return momentum, investors’ inattention, and limits to arbitrage. We find that this return predictability is mainly driven by investors’ limited information-processing capacity, and the evidence of predictability is stronger among low-localized firms with highly complicated business structures.

Detecting the lead–lag effect in stock markets: definition, patterns, and investment strategies

Human activities widely exhibit a power-law distribution. Considering stock trading as a typical human activity in the financial domain, the first aim of this paper is to validate whether the well-known power-law distribution can be observed in this activity. Interestingly, this paper determines that the number of accumulated lead–lag days between stock pairs meets the power-law distribution in both the U.S. and Chinese stock markets based on 10 years of trading data. Based on this finding this paper adopts the power-law distribution to formally define the lead–lag effect, detect stock pairs with the lead–lag effect, and then design a pure lead–lag investment strategy as well as enhancement investment strategies by integrating the lead–lag strategy into classic alpha-factor strategies. Tests conducted on 20 different alpha-factor strategies demonstrate that both perform better than the selected benchmark strategy and that the lead–lag strategy provides useful signals that significantly improve the performance of basic alpha-factor strategies. Our results therefore indicate that the lead–lag effect may provide effective information for designing more profitable investment strategies.

SPECTRAL FINANCIAL ECONOMETRICS

We survey the literature on spectral regression estimation. We present a cohesive framework designed to model dependence on frequency in the response of economic time series to changes in the explanatory variables. Our emphasis is on the statistical structure and on the economic interpretation of time-domain specifications needed to obtain horizon effects over frequencies, over scales, or upon aggregation. To this end, we articulate our discussion around the role played by lead-lag effects in the explanatory variables as drivers of differential information across horizons. We provide perspectives for future work throughout.

Lead-lag grey forecasting model in the new community group buying retailing

Community group buying is the preferred shopping method for many consumers during the COVID-19 epidemic in China. This new Internet pre-sale mode has a lead- lag effect and non-linear characteristics. Therefore, studying the prediction of community group buying has practical significance. In this paper, the lead-lag term of vector autoregressive model is coupled with the nonlinear term in the GM (1, 1, tα) model. Then, lead-lag factor τ and nonlinear factor tμ are added in the grey modelling process, and a novel hybrid grey forecasting model of lead-lag and nonlinearity (GAPM) is established by means of optimising the background value. Firstly, the least squares method and derivation method are used to obtain the model parameter values and time response expressions, and the stability of GAPM is studied on the basis of condition number theory of matrices. Secondly, the grey evolutionary algorithm is introduced to obtain the optimal solution of the background value and nonlinear factor, and the weighted geometric average weakening buffer operator is used to preprocess the original data. Finally, the new model is used to fit and predict the daily sales of Orange Optimization Company in Wuhan, China for two consecutive months of community group buying. Results show that the model has good stability and high prediction accuracy. Compared with the other five common grey models that do not consider the lead-lag influence of the system, GAPM has better predictive ability. Therefore, the introduction of the new model constructed by using the lead-lag factor is reasonable and effective in the case of community group buying with lead-lag effect.

Do collective emotions drive bitcoin volatility? A triple regime-switching vector approach

In this paper, we build an empirical specification that helps to explain bitcoin volatility and to characterize phases of the bitcoin bubble using information derived from investors’ emotions and sentiment that captures investment intentions and investors’ aversion to risk. To this end, we investigated the bilateral relations between bitcoin volatility and investor emotions between 2018 and 2021, a period characterized by significant changes in bitcoin prices as well as wide disparities in investor emotions, especially in the context of the ongoing COVID-19 pandemic. The study was based on a linear and nonlinear Vector Autoregressive (VAR) model that we applied to data related to bitcoin prices and market sentiment as expressed by the Fear and Greed index. Overall, our results evince the key role played by collective emotions in the formation and collapse of the bitcoin bubble. Two findings in particular stand out. First, our model shows significant time-varying lead-lag effects between bitcoin volatility and investor sentiment that come into play bilaterally and help to characterize the dynamics of bitcoin volatility. Second, these interactions exhibit asymmetry and nonlinearity as the sign and size of collective emotions (resp. bitcoin volatility) vary with the regime and market state under consideration (calm state versus period of bubble formation, etc.). In other words, the power of sentiment has a time-varying effect on the market. Indeed, in the first regime (“calm state”), where bitcoin volatility is relatively low and the market shows evidence of stability, collective emotions have a negative impact on bitcoin volatility, prompting a stabilizing strength. However, in the second regime (“bubble formation”), the effect of emotions turns significantly positive as investors gradually become less fearful and more reassured, which can simultaneously increase volatility and destabilize the market. Finally, in the third regime (“bubble collapse”), when bitcoin reaches a high level of value and experiences impressive volatility excess, the effect of emotions again turns negative, resulting in further switching behavior that pushes investor action to provoke a bitcoin price correction, moving it toward a new state of stability. Our conclusion helps improve predictions of bitcoin price dynamics informed by the information provided by investor emotions.

Concept links and return momentum

Unlike traditional asset categories (e.g., industry classifications) that are generally defined clearly, some groups of stocks are tied to certain loosely defined “concepts” (e.g., e-commerce). When investors find it difficult to analyze ambiguous concept-oriented information, information diffuses slowly, creating “concept momentum”. Based on unique concept data in the Chinese stock market, this study constructs a concept-momentum strategy that involves buying stocks from past winning concepts and selling stocks from past losing concepts, which can generate pronounced abnormal returns. Neither risk factors, firm-level momentum, nor industry-level momentum can explain concept momentum. Furthermore, we find that both the underreaction and cross-stock lead-lag effect channels can cause slow information diffusion and drive concept momentum. Moreover, the concept momentum effect is stronger for relatively ambiguous concepts, for concepts that attract less investor attention, and following high-sentiment periods.

Financial time series forecasting with multi-modality graph neural network

Financial time series analysis plays a central role in hedging market risks and optimizing investment decisions. This is a challenging task as the problems are always accompanied by multi-modality streams and lead-lag effects. For example, the price movements of stock are reflections of complicated market states in different diffusion speeds, including historical price series, media news, associated events, etc. Furthermore, the financial industry requires forecasting models to be interpretable and compliant. Therefore, in this paper, we propose a multi-modality graph neural network (MAGNN) to learn from these multimodal inputs for financial time series prediction. The heterogeneous graph network is constructed by the sources as nodes and relations in our financial knowledge graph as edges. To ensure the model interpretability, we leverage a two-phase attention mechanism for joint optimization, allowing end-users to investigate the importance of inner-modality and inter-modality sources. Extensive experiments on real-world datasets demonstrate the superior performance of MAGNN in financial market prediction. Our method provides investors with a profitable as well as interpretable option and enables them to make informed investment decisions.

Are Cryptocurrencies and African stock markets integrated?

The finance literature provides evidence that the cryptocurrency market is integrating with other markets. This may induce increased investor participation with the chance of excessive liquidity in the cryptocurrency market, which can impair financial stability should there be shocks to the cryptocurrency market. This provides grounds to uncover the degree of integration between cryptocurrencies and African stock markets, which has been completely untouched using wavelet-based methods and frequency domain spillover index. Such knowledge is critical since shocks to the cryptocurrency market may have rippling effects on African stock markets. Findings suggest low degrees of integration between the markets at higher frequencies, which grows stronger at medium frequencies and perfectly integrates at lower frequencies. The implication is that stock markets in Africa are highly exposed to cryptocurrency market disruptions from the medium-term, and international investors seeking to hedge their price risk in African stock markets using cryptocurrencies may have to look at the short term. We find cryptocurrencies as significant spillover transmitters across frequencies. The lead (lag) effects are time-varying and heterogeneous, showing no particular market as a leader, suggesting arbitrage opportunities for investors.

Dynamic patterns of daily lead-lag networks in stock markets

The lead-lag relationship between stocks is an interesting phenomenon, which has been empirically seen to widely exist in stock markets. This paper aims to discover the dynamic patterns of the daily lead-lag relationships between stock pairs, to detect the features of the discovered dynamic patterns, and to explore which factors significantly affect the emergence of the feature. To this end, a series of statistical analyses is conducted to find that the (longest) successive lead-lag days satisfy a power-law distribution in the two mainland stock markets in China, which answers the question regarding the dynamic pattern. Note that the heavy tail of the power-law distribution is the core of the discovered dynamic pattern. A formal and solid definition of the lead-lag effect is provided by statistical testing, and then the corresponding detection method is designed and applied to obtain the heavy tail. Finally, an empirical study of the detected stocks with lead-lag effect is further conducted via an exponential random graph model (ERGM). Our work adds new knowledge to the lead-lag phenomenon in the financial domain, provides a formal definition of the lead-lag effect and proposes a new detection method benefiting future studies on the lead-lag relationship in financial markets. It further contributes to the existing relevant literature by a deep understanding of which factors cause the emergence of the power-law distribution discovered.

The interrelationship between order flow, exchange rate, and the role of American economic news

The aim of this study is to investigate the comovement and interconnection between US economic news in seven different categories and the exchange rate of the US dollar against the currencies of seven developed countries, with the corresponding order flows. Our empirical investigation is based on using the Spearman correlation method to analyze the correlation between the economic news and exchange rate using order flows before, during, and after the financial news announcement. We use wavelet transform analysis to assess the comovement in the time–frequency domains. The bivariate outcomes demonstrate that changes in economic news impact price changes for only a short time, or even no time at all. Moreover, for longer time periods, the order flow is shown to have stronger coherency than economic news and a steady comovement against the exchange rate. Wavelet coherency reveals that the lead-lag effect of order flow changes on price changes starts in the medium frequency bands to low-frequency bands during the entire sample period. In addition, economic news weakens the correlation between exchange rate and order flow in the short run, which means that the combination of economic news and order flow can achieve a higher degree of consistency of exchange rate. The implication of this study is that macro policy makers, import and export enterprises, foreign exchange investors and exchange fund managers can predict the future exchange rate based on the order flow and choose hedging measures under different objectives.

The connectedness of house price affordability (HPA) and rental price affordability (RPA) measures

PurposeThe purpose of this paper is to examine the relationship between house price affordability (HPA) and rental price affordability (RPA) in New Zealand. The cointegration of HPA and RPA is of particular focus given rising house prices and rising rents.Design/methodology/approachThe study examines the lead-lad correlation between HPA and RPA. The method uses a generalised least square technique and the development of an ordinary least squares model.FindingsThe study shows that there is an existence of cointegration and unidirectional statistical causality effects between HPA and RPA across 11 regions in New Zealand. Furthermore, Auckland, Wellington and Canterbury are the three regions in which the results detect the most extreme effects amongst HPA and RPA compared to other places in the country. Extended empirical work shows interesting results that there are lead-lag effects of HPA and RPA on each other and on mortgage rates at the national scale. These effects are consistent for both methods but are changed at individual lead-lag variables and amongst different regions.Originality/valueThe study empirically provides useful insight for both academia and practitioners. Particularly in examining the long-run effects, cointegration and forecasting of the volatile interactions between HPA and RPA.

The Identification and Prediction in Abundance Variation of Atlantic Cod via Long Short-Term Memory With Periodicity, Time–Frequency Co-movement, and Lead-Lag Effect Across Sea Surface Temperature, Sea Surface Salinity, Catches, and Prey Biomass From 1919 to 2016

The population of Atlantic cod significantly contributes to the prosperity of fishery production in the world. In this paper, we quantitatively investigate the global abundance variation in Atlantic cod from 1919 to 2016, in favor of spatiotemporal interactions over manifold impact factors at local observation sites, and propose to explore the predictive mechanism with the help of its periodicity, time–frequency co-movement, and lead-lag effects, via long short-term memory (LSTM). We first integrate evidences yielded from wavelet coefficients, to suggest that the abundance variation potentially follows a 36-year major cycle and 24-year secondary cycle at the time scales of 55 years and 37 years. We further evaluate the responses of Atlantic cod abundance to the external impact factors, including sea surface temperature (SST), catches, prey biomass, and sea surface salinity (SSS), in aid of the wavelet coherence and phase difference, which allows us to identify the dominantly correlative factors and capture the leading roles along the time domain and then divide the responses around the recent 60 years into three stages: before 1985, 1985–1995, and after 1995. At the first stage, the reason for the decline in abundance could be mainly attributed to the rapid rise of fish catches. At the second stage, the impact of SST and SSS also provides significant indices, besides overfishing; meanwhile, the mortality of primary producers and forced migration of fish species indirectly cause the decline. At the third stage, warming SST and growing SSS directly led to the decrease of abundance. Finally, we establish one ensemble of LSTM-SAE architecture to comprehensively reflect the predictive patterns at each stage. It has been demonstrated from experimental results that the models behaved better when intentionally feeding with the dominantly correlative multivariate inputs, instead of either all factors or only the abundance. The proposed scheme provides opportunities to symmetrically identify the underlying predictive attributes of Atlantic cod abundance and potentially perform as the quantitative references in reasonably making fishing decision. With the rapid development in deep learning capabilities, it is hopeful to expect better predictions of the responses to global changes, not only for Atlantic cod but also for other fish species and the ecosystem as a whole.

Does the lead-lag effect exist in stock markets?

ABSTRACT This paper investigates the lead-lag effect from a complex network perspective. We first detect the lead-lag effect between individual stocks based on CSI 300 index in Chinese stock market and then employed a stochastic actor-oriented model to investigate the interrelationship between the detected lead-lag network and stocks’ characteristics. The main result is that market capitalization, trading volume and financial performance are significant driving factors that form the lead-lag relationship.

Economic policy uncertainty and hotel occupancy: the mediating effect of consumer sentiment

PurposeThe purpose of this paper is twofold. First, the study examines the prolonged effect of policy-related economic uncertainty on hotel operating performance, particularly the room demand (occupancy). Second, the study attempts to explain why occupancy drops when the perceived economic uncertainty is high by studying the mediating effect of consumer sentiment in the relationship between economic policy uncertainty and hotel demand.Design/methodology/approachThis quantitative study uses secondary data – US economic policy uncertainty (EPU) index, University of Michigan's index of consumer sentiment (ICS), and property-level hotel operating data from three states of the US – California, Florida and New York. Data were analyzed using random effect regression and structural equation modeling. Robustness tests were conducted to enhance the reliability of the research findings.FindingsRandom-effects regression analysis reveals that policy-related economic uncertainty has a negative and lead-lag effect on hotel occupancy, average daily rate and revenue per available room (RevPAR). Structural equation modeling results show that the relationship between economic policy uncertainty and hotel occupancy is significantly mediated by consumer sentiment. Robustness test results support the findings from the main analysis.Practical implicationsThis study offers valuable implications for the hotel professionals in regard to anticipating the economic impact of policy-related uncertainty on hotel industry and understanding how consumer sentiment affects demand at such crises times. Moreover, the study suggests potential course of actions to deal with declining room demand at times of uncertainty.Originality/valueThis empirical study explores how economic policy uncertainty affects hotel performance at the property level and explains the mediating effect of consumer sentiment on hotel room demand. The study provides a first-hand evidence of how consumer sentiment relates to the perception of economic uncertainty and leads to decline in consumer demand. In that regard, findings of the study have valuable implications for hospitality industry practitioners and relevant policymakers.

Labor Market Networks and Asset Returns

This paper proposes a measure of labor market connectivity based on the similarity in the composition of occupational knowledge characteristics across industries and provides evidence of return predictability in the cross-section of industries that are connected through the labor market. In long-short portfolios, an industry's return is strongly predicted by the past return of its labor-market-connected neighboring industries with an annualized return of up to 9%, which is not explained by established asset pricing models. The return predictability remains significant after controlling for the supply chain momentum as well as the industry lead-lag effect, and is concentrated in stocks with higher arbitrage costs and higher ownership of uninformed investors. We find similar predictive relations for the labor productivity, wages, employment, and profitability of labor connected industries. Our findings are consistent with positive spillover of productivity shocks among industries that are connected through the labor market. Informational frictions, costly arbitrage, and investors' limited attention magnify the delayed response of stock prices to the spillover of labor productivity shocks, which results in the observed return predictability.

How to Identify Varying Lead–Lag Effects in Time Series Data: Implementation, Validation, and Application of the Generalized Causality Algorithm

This paper develops the generalized causality algorithm and applies it to a multitude of data from the fields of economics and finance. Specifically, our parameter-free algorithm efficiently determines the optimal non-linear mapping and identifies varying lead–lag effects between two given time series. This procedure allows an elastic adjustment of the time axis to find similar but phase-shifted sequences—structural breaks in their relationship are also captured. A large-scale simulation study validates the outperformance in the vast majority of parameter constellations in terms of efficiency, robustness, and feasibility. Finally, the presented methodology is applied to real data from the areas of macroeconomics, finance, and metal. Highest similarity show the pairs of gross domestic product and consumer price index (macroeconomics), S&P 500 index and Deutscher Aktienindex (finance), as well as gold and silver (metal). In addition, the algorithm takes full use of its flexibility and identifies both various structural breaks and regime patterns over time, which are (partly) well documented in the literature.

Epidemiology of Coronavirus COVID-19: Forecasting the Future Incidence in Different Countries.

This paper forecasts the future spread of COVID-19 by exploiting the identified lead-lag effects between different countries. Specifically, we first determine the past relation among nations with the aid of dynamic time warping. This procedure allows an elastic adjustment of the time axis to find similar but phase-shifted sequences. Afterwards, the established framework utilizes information about the leading country to predict the Coronavirus spread of the following nation. The presented methodology is applied to confirmed Coronavirus cases from 1 January 2020 to 28 March 2020. Our results show that China leads all other countries in the range of 29 days for South Korea and 44 days for the United States. Finally, we predict a future collapse of the healthcare systems of the United Kingdom and Switzerland in case of our explosion scenario.

The Leaders, the Laggers, and the “Vulnerables”

We examine the lead-lag effect between the large and the small capitalization financial institutions by constructing two global weekly rebalanced indices. We focus on the 10% of stocks that “survived” all the rebalancings by remaining constituents of the indices. We sort them according to their systemic importance using the marginal expected shortfall (MES), which measures the individual institutions’ vulnerability over the market, the network based MES, which captures the vulnerability of the risks generated by institutions’ interrelations, and the Bayesian network based MES, which takes into account different network structures among institutions’ interrelations. We also check if the lead-lag effect holds in terms of systemic risk implying systemic risk transmission from the large to the small capitalization, concluding a mixed behavior compared to the index returns. Additionally, we find that all the systemic risk indicators increase their magnitude during the financial crisis.