Search

While attention is a predictor for digital asset prices, and jumps in Bitcoin prices are well-known, we know little about its alternatives. Studying high frequency crypto data gives us the unique possibility to confirm that cross market digital asset returns are driven by high frequency jumps clustered around black swan events, resembling volatility and trading volume seasonalities. Regressions show that intra-day jumps significantly influence end of day returns in size and direction. This provides fundamental research for crypto option pricing models. However, we need better econometric methods for capturing the specific market microstructure of cryptos. All calculations are reproducible via the quantlet.com technology.

The centrality of the United States in the global financial system is taken for granted, but its response to recent political and epidemiological events has suggested that China now holds a comparable position. Using minute-by-minute data from 2012 to 2020 on the financial performance of twelve country-specific exchange-traded funds, we construct daily snapshots of the global financial network and analyze them for the centrality and connectedness of each country in our sample. We find evidence that the U.S. was central to the global financial system into 2018, but that the U.S.-China trade war of 2018-2019 diminished its centrality, and the Covid-19 outbreak of 2019-2020 increased the centrality of China. These indicators may be the first signals that the global financial system is moving from a unipolar to a bipolar world.

Submitted by CVPortalegre Deposit (cv_ipportalegre@cienciavitae.pt) on 2021-11-12T23:21:13Z No. of bitstreams: 2 Intraday_Volatility_Spillovers_among_European_Financial_Markets_during_COVID-19.pdf: 5167097 bytes, checksum: c3b6c071680fe1dfbfe1db4e9b08fdd7 (MD5) package.zip: 4391214 bytes, checksum: 0a7133008235654c1ce4b6ea81e137cd (MD5) Approved for entry into archive by Catarina Dias (catarina.dias@ipportalegre.pt) on 2021-11-12T23:37:53Z (GMT) No. of bitstreams: 2 package.zip: 4391214 bytes, checksum: 0a7133008235654c1ce4b6ea81e137cd (MD5) Intraday_Volatility_Spillovers_among_European_Financial_Markets_during_COVID-19.pdf: 5167097 bytes, checksum: c3b6c071680fe1dfbfe1db4e9b08fdd7 (MD5) Approved for entry into archive by Catarina Dias (catarina.dias@ipportalegre.pt) on 2021-11-12T23:38:07Z (GMT) No. of bitstreams: 2 package.zip: 4391214 bytes, checksum: 0a7133008235654c1ce4b6ea81e137cd (MD5) Intraday_Volatility_Spillovers_among_European_Financial_Markets_during_COVID-19.pdf: 5167097 bytes, checksum: c3b6c071680fe1dfbfe1db4e9b08fdd7 (MD5) Made available in DSpace on 2021-11-12T23:38:07Z (GMT). No. of bitstreams: 2 package.zip: 4391214 bytes, checksum: 0a7133008235654c1ce4b6ea81e137cd (MD5) Intraday_Volatility_Spillovers_among_European_Financial_Markets_during_COVID-19.pdf: 5167097 bytes, checksum: c3b6c071680fe1dfbfe1db4e9b08fdd7 (MD5) Previous issue date: 2021-01-05 N/A

We studied (i) the volatility feedback effect, defined as the relationship between contemporaneous returns and the market-based volatility, and (ii) the leverage effect, defined as the relationship between lagged returns and the current market-based volatility. For our analysis, we used daily measures of volatility estimated from high frequency data to explain volatility changes over time for both the S&amp, P500 and FTSE100 indices. The period of analysis spanned from January 2000 to June 2017 incorporating various market phases, such as booms and crashes. Based on the estimated regressions, we found evidence that the returns of S&amp, P500 and FTSE100 indices were well explained by a specific group of realized measure estimators, and the returns negatively affected realized volatility. These results are highly recommended to financial analysts dealing with high frequency data and volatility modelling.

What drives the strong reaction of financial markets to central bank communication on the days of policy decisions? We highlight the role of two factors that we identify from highfrequency monetary surprises: news on future macroeconomic conditions (Delphic shocks) and news on future monetary policy shocks (Odyssean shocks). These two shocks move the yield curve in the same direction but have opposite effects on financial conditions and macroeconomic expectations. A drop in future interest rates that is associated with a negative Delphic (Odyssean) shock is perceived as being contractionary (expansionary). These offsetting effects can explain why central bank communication leads to a strong reaction of the yield curve together with a weak reaction by inflation expectations or stock prices. The two shocks also have different impacts on macroeconomic outcomes, such that central bankers cannot infer the degree of stimulus they provide by looking at the mere reaction of the yield curve. However, changes in their communication policy can influence the way markets predominantly understand communication about future interest rates.

Forecasting volatility models typically rely on either daily or high frequency (HF) data and the choice between these two categories is not obvious. In particular, the latter allows to treat volatility as observable but they suffer from many limitations. HF data feature microstructure problem, such as the discreteness of the data, the properties of the trading mechanism and the existence of bid-ask spread. Moreover, these data are not always available and, even if they are, the asset’s liquidity may be not sufficient to allow for frequent transactions. This paper considers different variants of these two family forecasting-volatility models, comparing their performance (in terms of Value at Risk, VaR) under the assumptions of jumps in prices and leverage effects for volatility. Findings suggest that daily-data models are preferred to HF-data models at 5% and 1% VaR level. Specifically, independently from the data frequency, allowing for jumps in price (or providing fat-tails) and leverage effects translates in more accurate VaR measure.

The problem of estimation of realized correlation, which is analogous to realized covariance, is compounded by effects of market microstructure, noise, and asynchronous trading. Various methods have been proposed to decrease the biases, but they require assumptions to be made that may be unrealistic. This paper argues that the inherent data problems make precise point identification of realized correlation difficult, but identification bounds in the spirit of Manski (1995) can be derived. These identification bounds allow for a more robust approach to inference, especially when the realized correlation is used for estimating other risk measures. We forecast the identification bounds using the HAR model of Corsi (2003) using data during the year of onset of the credit crisis, and find that the bounds provide good predictive coverage of the realized correlation for both one step and ten step forecasts, even in volatile periods.

Forecasting-volatility models typically rely on either daily or high frequency (HF) data and the choice between these two categories is not obvious. In particular, the latter allows to treat volatility as observable but they suffer of many limitations. HF data feature microstructure problem, such as the discreteness of the data, the properties of the trading mechanism and the existence of bid-ask spread. Moreover, these data are not always available and, even if they are, the asset’s liquidity may be not sufficient to allow for frequent transactions. This paper considers different variants of these two family forecasting-volatility models, comparing their performance (in terms of Value at Risk, VaR) under the assumptions of jumping prices and leverage effects for volatility. Findings suggest that GARJI model provides more accurate VaR measures for the S&P 500 index than RV models. Furthermore, the assumption of conditional normality is shown to be not sufficient to obtain accurate risk measures even if jump contribution is provided. More sophisticated models might address this issue, improving VaR results.

In this paper, the relationship between macroeconomic fundamentals and asset prices is explored by estimating the impact of macroeconomic announcements in the Brazilian futures market. Using intraday data from October 2008 to January 2011, results show that external macroeconomic announcements dominate price changes in the Foreign Exchange and Ibovespa futures markets, while the impact of the domestic ones is mainly restricted to Interest Rate futures contracts. There is also evidence that price reactions are conditional on the state of the economy.

The estimation of the impact of macroeconomic announcements in the Brazilian futuresmarkets is used to uncover the relationship between macroeconomic fundamentals andasset prices. Using intraday data from October 2008 to January 2011, we find thatexternal macroeconomic announcements dominate price changes in the ForeignExchange and Ibovespa futures markets, while the impact of the domestic ones ismainly restricted to Interest Rate futures contracts. We additionally propose aninvestment strategy based on the conditional price reaction of each market that showedpromising results in an out-of-sample study, where we are able to correctly identifyreturns’ signals, conditional on the surprise’s signal, in approximately 70% of the cases.Finally, we provide evidence that price reactions are conditional on the state of theeconomy and document the impact on volume and bid-ask spreads.

This study aims to analyze price discovery in the foreign exchange market in Brazil and indicate which market (spot or futures) adjusts more quickly to the arrival of new information. Using high frequency data from January/2008 to June/2013, we estimate the following price discovery metrics: Information Share (IS), Component Share (CS) and Price Discovery Efficiency Loss (PDEL). We find that futures market dominates price discovery since it responds for 66.2% of the variation in the fundamental price shock and for 97.4% of the fundamental price composition. In a dynamic perspective, futures market is also more efficient since, when markets are subjected to a shock in the fundamental price, it is faster to recover to equilibrium. We also investigated whether results are robust to sub-samples. When we break in sub-samples by semester, price discovery figures confirm futures dominance in all sub-samples. Spot market offer-demand disequilibrium, central bank interventions and external institutional investors' pressure in the futures market emerge as potential explanatory factors to the variation in price discovery metrics between semesters.

Policy impact studies often suffer from endogeneity problems. Consider the case of the ECB Securities Markets Programme: If Eurosystem interventions were triggered by sudden and strong price deteriorations, looking at daily price changes may bias downwards the correlation between yields and the amounts of bonds purchased. Simple regression of daily changes in yields on quantities often give insignificant or even positive coefficients and therefore suggest that SMP interventions have been ineffective, or worse counterproductive. We use high frequency data on purchases of the ECB Securities Markets Programme and sovereign bond quotes to address the endogeneity issues. We propose an econometric model that considers, simultaneously, first and second conditional moments of market price returns at daily and intradaily frequency. We find that SMP interventions succeeded in reducing yields and volatility of government bond segments of the countries under the programme. Finally, the new econometric model is broadly applicable to market intervention studies.

Copulae sind ein hilfreiches Werkzeug um die Abhängigkeitsstrukturen zwischen Risikofaktoren in Wirtschaft und Finanzwesen zu modellieren und zu quantifizieren. Dank technologischen Entwicklungen der letzten Jahrzehnte, können hoch-frequentierte Daten mit innertäglichen Beobachtungen genutzt werden um die Covarianz zwischen Wertpapieren zu schätzen. Aus dieser Schätzung und zusätzlichen Annahmen über die Randverteilungen werden in dieser Arbeit Realized Copula Modelle an Hand von Aktienkursen auf verschiedenen Wegen geschätzt und untersucht. Basierend auf hoch-frequentierten Daten werden drei Schätzer für den Copula Parameter angewendet und durch die Nutzung von täglichen Renditen wird ein Maximum Likelihood Schätzer bestimmt. Dabei werden Eigenschaften erfasst, wo die multivariate Verteilung an ihre Grenzen kommt, wie zum Beispiel im Falle von Heavy-tailed Verteilungen und asymetrischen Abhängigkeiten. Copulae ermöglichen es die Randverteilungen seperat von der gemeinsamen Verteilung zu bestimmen. Die unterschiedlich geschätzten Copula parameter werden verglichen und die resultierenden Copula Modelle bezüglich ihrer Angepasstheit an die Daten und ihrer Vorhersage von Verlusten untersucht. Die modellierte Abhängigkeit ändert sich über die Zeit und unterscheidet sich je nach verfügbarer Kovarianzschätzung und Ziehungshäufigkeit der innertages Stichproben. Dabei werden die Realized Kernel Schätzer, bei sehr häufiger Ziehung, zum Teil bei der Schätzung der Copula Parameter genutzt. Einige Parameterschätzer liefern präzisere Quantilsschätzungen der Verlustverteilung als andere, was durch eine Value-at-Risk Schätzung und durch backtesting gezeigt wird. Abstract For modeling and quantifying the dependence structure between financial risk factors like stock returns, copula models are a useful instrument. Abstract Due to technological growth and improvements of the last decades, high-frequency data with intra-daily observations can be used to estimate covariance of assets. From these estimates and assumed marginal distribution, realized copula models are constructed and analyzed using real price data. Abstract Based on high-frequency data, three different estimation methods are applied and in addition daily stock return are used to get Maximum Likelihood estimates. All estimators are implemented and the resulting copula models are compared according to their ability to describe the true multivariate distribution and forecast the losses. Abstract The modeled dependence parameter are time-varying and differ according to the estimated covariance matrix and to the sampling frequency. Moreover, all estimation steps are done based on data with high sampling frequencies like seconds and lower frequency of 5 minutes. The realized kernel estimator is used to cope with market microstructure effects and the sensitivity to this noise can be reflected in the Value-at-Risk performance.

International audience; The spread of the global financial crisis of 2008/2009 was rapid, and impacted the functioning and the performance of financial markets. Due to the importance of this phenomenon, this study aims to explain the impact of the crisis on stock market behavior and interdependence through the study of the intraday volatility transmission. This paper investigates the patterns of linkage dynamics among three European stock markets—France, Germany, and the UK—during the global financial crisis, by analyzing the intraday dynamics of linkages among these markets during both calm and turmoil phases. We apply a VAR-EGARCH (Vector Autoregressive Exponential General Autoregressive Conditional Heteroscedasticity) framework to high frequency five-minute intraday returns on selected representative stock indices. We find evidence that interrelationship among European markets increased substantially during the period of crisis, pointing to an amplification of spillovers. In addition, during this period, French and UK markets herded around German market, possibly explained by behavior factors influencing the stock markets on or near dates of extreme events. Germany was identified as the hub of financial and economic activity in Europe during the period of study. These findings have important implications for both policymakers and investors by contributing to better understanding the transmission of financial shocks in Europe.

This paper analyses the long-memory properties of a high-frequency financial time series dataset. It focuses on temporal aggregation and other features of the data, and how they might affect the degree of dependence of the series. Fractional integration or I(d) models are estimated with a variety of specifications for the error term. In brief, we find evidence that a lower degree of integration is associated with lower data frequencies. In particular, when the data are collected every 10 minutes there are several cases with values of d strictly smaller than 1, implying mean-reverting behaviour; however, for higher data frequencies the unit root null cannot be rejected. This holds for all four series examined, namely Open, High, Low and Last observations for the US dollar / British pound spot exchange rate and for different sample periods.

This paper analyses the long-memory properties of a high-frequency financial time series dataset. It focuses on temporal aggregation and other features of the data, and how they might affect the degree of dependence of the series. Fractional integration or I(d) models are estimated with a variety of specifications for the error term. In brief, we find evidence that a lower degree of integration is associated with lower data frequencies. In particular, when the data are collected every 10 minutes there are several cases with values of d strictly smaller than 1, implying mean-reverting behaviour; however, for higher data frequencies the unit root null cannot be rejected. This holds for all four series examined, namely Open, High, Low and Last observations for the US dollar / British pound spot exchange rate and for different sample periods.

This dissertation comprises of three stand-alone research papers, all considering the use of high frequency financial data for financial market risk measurement. The first chapter considers the extraction of liquidity information from the intraday limit order book to enhance the daily market risk measures for the dimension of liquidity risk. While liquidity risk in trading is understood to be important for practitioners, it is not well understood how it should be quantified and included in estimating market risks. The second and third chapter contribute to the ongoing research on realized measures in the multivariate context. Chapter two provides some new ideas about how to characterise the ex-post dependence, which we term as "realized dependence", between two or more assets using intraday data. Chapter three describes a new approach via partial identification to deal with the biases that currently plague the estimation of realized covariance and correlation.

We provide empirical evidence of volatility forecasting in relation to asymmetries present in the dynamics of both return and volatility processes. Using recently-developed methodologies to detect jumps from high frequency price data, we estimate the size of positive and negative jumps and propose a methodology to estimate the size of jumps in the quadratic variation. The leverage effect is separated into continuous and discontinuous effects, and past volatility is separated into “good” and “bad”, as well as into continuous and discontinuous risks. Using a long history of the S & P500 price index, we find that the continuous leverage effect lasts about one week, while the discontinuous leverage effect disappears after one day. “Good” and “bad” continuous risks both characterize the volatility persistence, while “bad” jump risk is much more informative than “good” jump risk in forecasting future volatility. The volatility forecasting model proposed is able to capture many empirical stylized facts while still remaining parsimonious in terms of the number of parameters to be estimated.

In dieser Dissertation befassen wir uns mit ökonometrischen Modellen und empirischen Eigenschaften von Intra-Tages (Hochfrequenz-) Aktienmarktdaten. Der Fokus liegt hierbei auf der Analyse des Einflusses, den die Veröffentlichung von Wirtschaftsnachrichten auf die Aktienmarktaktivität hat, der Vorhersage der Geld-Brief-Spanne sowie der Modellierung von Volatilitätsmaßen auf Intra-Tages-Zeitintervallen. Zunächst quantifizieren wir die Marktreaktionen auf Marktneuigkeiten innerhalb eines Handelstages. Zu diesem Zweck benutzen wir linguistisch vorab bearbeitete Unternehmensnachrichtendaten mit Indikatoren über die Relevanz, Neuheit und Richtung dieser Nachrichten. Mit einem VAR Modell für 20-Sekunden Marktdaten der London Stock Exchange weisen wir durch Nachrichten hervorgerufene Marktreaktionen in Aktienkursrenditen, Volatilität, Handelsvolumina und Geld-Brief-Spannen nach. In einer zweiten Analyse führen wir ein long memory autoregressive conditional Poisson (LMACP)-Modell zur Modellierung hoch-persistenter diskreter positivwertiger Zeitreihen ein. Das Modell verwenden wir zur Prognose von Geld-Brief-Spannen, einem zentralen Parameter im Aktienhandel. Wir diskutieren theoretische Eigenschaften des LMACP-Modells und evaluieren rollierende Prognosen von Geld-Brief-Spannen an den NYSE und NASDAQ Börsenplätzen. Wir zeigen, dass Poisson-Zeitreihenmodelle in diesem Kontext signifikant bessere Vorhersagen liefern als ARMA-, ARFIMA-, ACD- und FIACD-Modelle. Zuletzt widmen wir uns der optimalen Messung von Volatilität auf kleinen 20 Sekunden bis 5 Minuten Zeitintervallen. Neben der Verwendung von realized volatility-Ansätzen konstruieren wir Volatilitätsmaße durch Integration von spot volatility-Schätzern, sodass auch Beobachtungen außerhalb der kleinen Zeitintervalle in die Volatilitätsschätzungen eingehen. Ein Vergleich der Ansätze in einer Simulationsstudie zeigt, dass Volatilitätsmaße basierend auf spot volatility-Schätzern den RMSE minimieren., In this thesis we present econometric models and empirical features of intra-daily (high frequency) stock market data. We focus on the measurement of news impacts on stock market activity, forecasts of bid-ask spreads and the modeling of volatility measures on intraday intervals. First, we quantify market reactions to an intraday stock-specific news flow. Using pre-processed data from an automated news analytics tool we analyze relevance, novelty and direction signals and indicators for company-specific news. Employing a high-frequency VAR model based on 20 second data of a cross-section of stocks traded at the London Stock Exchange we find distinct responses in returns, volatility, trading volumes and bid-ask spreads due to news arrivals. In a second analysis we introduce a long memory autoregressive conditional Poisson (LMACP) model to model highly persistent time series of counts. The model is applied to forecast quoted bid-ask spreads, a key parameter in stock trading operations. We discuss theoretical properties of LMACP models and evaluate rolling window forecasts of quoted bid-ask spreads for stocks traded at NYSE and NASDAQ. We show that Poisson time series models significantly outperform forecasts from ARMA, ARFIMA, ACD and FIACD models in this context. Finally, we address the problem of measuring volatility on small 20 second to 5 minute intra-daily intervals in an optimal way. In addition to the standard realized volatility approaches we construct volatility measures by integrating spot volatility estimates that include information on observations outside of the intra-daily intervals of interest. Comparing the alternative volatility measures in a simulation study we find that spot volatility-based measures minimize the RMSE in the case of small intervals.

We elaborate economic explanations for the time-varying risk of month, quarter and year base load electricity forward contracts traded on the Nord Pool Energy Exchange from January 2006 to March 2010. Daily risk quantities are generated by decomposing realized volatility in its continuous and discontinuous jump component. First, we analyze the relation between volatility and trading activity. Coherent with existing studies we find that the driving factor of the relation between continuous variation and trading activity is the number of trades. New insights are obtained by considering the relation between jump factor and trading activity. Our results indicate that the number of trades and absolute order imbalance, which can be explicitly measured in our dataset, are positively related to the jump factor, a result in line with theoretical models. Second, we study unscheduled news announcements causing high volatilities. For this, a unique dataset of urgent market messages (UMMs), published by the Nord Pool Energy Exchange, is created. We extract relevant unscheduled UMMs, here failures, from both transmission system operators (TSOs) and market participants (MPs), and measure their impact over varying event windows. We find that certain unscheduled TSO/MP-UMMs have a significant impact on continuous variation, especially when they are published close to maturity, their content refers to a rare and extreme event or the contract is a month forward. The analysis also provides economic evidence for the occurrence of price jumps.

This paper analyses the long-memory properties of high frequency financial time series. It focuses on temporal aggregation and the influence that this might have on the degree of dependence of the series. Fractional integration or I(d) models are estimated with a variety of specifications for the error term. In brief, we find evidence that a lower degree of integration is associated with lower data frequencies. In particular, when the data are collected every 10 minutes there are several cases with values of d strictly smaller than 1, implying mean-reverting behaviour. This holds for all four series examined, namely Open, High, Low and Last observations for the British pound/US dollar spot exchange rate. The second-named author gratefully acknowledges financial support from the Ministerio de Ciencia y Tecnología (ECO2008-03035 ECON Y FINANZAS, Spain) and from a PIUNA Project of the University of Navarra.

We propose a unified framework for estimating integrated variances and covariances based on simple OLS regressions, allowing for a general market microstructure noise specification. We show that our estimators can outperform in terms of the root mean squared error criterion the most recent and commonly applied estimators, such as the realized kernels of Barndorff-Nielsen, Hansen, Lunde & Shephard (2006), the two-scales realized variance of Zhang, Mykland & Ait-Sahalia (2005), the Hayashi & Yoshida (2005) covariance estimator, and the realized variance and covariance with the optimal sampling frequency chosen after Bandi & Russell (2005a) and Bandi & Russell (2005b). The power of our methodology stems from the fact that instead of trying to correct the realized quantities for the noise, we identify both the true underlying integrated moments and the moments of the noise, which are also estimated within our framework. Apart from being simple to implement, an important property of our estimators is that they are quite robust to misspecifications of the noise process.

This paper applies a non- and a semiparametric copula-based approach to analyze the first-order autocorrelation of returns in high frequency financial time series. Using the EUREX D3047 tick data from the German stock index, it can be shown that the temporal dependence structure of price movements is not always negatively correlated as assumed in the stylized facts in the finance literature. Depending on the sampling frequency, the estimated copulas exhibit some kind of overreaction phenomena and multiple tail dependence, revealing patterns similar to the compass rose.

This paper applies a non- and a semiparametric copula-based approach to analyze the first-order autocorrelation of returns in high frequency financial time series. Using the EUREX D3047 tick data from the German stock index, it can be shown that the temporal dependence structure of price movements is not always negatively correlated as assumed in the stylized facts in the finance literature. Depending on the sampling frequency, the estimated copulas exhibit some kind of overreaction phenomena and multiple tail dependence, revealing patterns similar to the compass rose.

This study analyzes the short-term dynamic spillovers between the futures returns on the DAX, the DJ Eurostoxx 50 and the FTSE 100. It also examines whether economic news is one source of international stock return co-movements. In particular, we test whether stock market interdependencies are attributable to reactions of foreign traders to public economic information. Moreover, we analyze whether cross-market linkages remain the same or whether they do increase during periods in which economic news is released in one of the countries. Our main results can be summarized as follows: (i) there are clear short term international dynamic interactions among the European stock futures markets; (ii) foreign economic news affects domestic returns; (iii) futures returns adjust to news immediately; (iv) announcement timing of macroeconomic news matters; (v) stock market dynamic interactions do not increase at the time of the release of economic news; (vi) foreign investors react to the content of the news itself more than to the response of the domestic market to the national news; and (vii) contemporaneous correlation between futures returns changes at the time of macroeconomic releases. JEL Classification: G14, G15

In this paper, we give an overview of the state-of-the-art in the econometric literature on the modeling of so-called financial point processes. The latter are associated with the random arrival of specific financial trading events, such as transactions, quote updates, limit orders or price changes observable based on financial high-frequency data. After discussing fundamental statistical concepts of point process theory, we review duration-based and intensity-based models of financial point processes. Whereas duration-based approaches are mostly preferable for univariate time series, intensity-based models provide powerful frameworks to model multivariate point processes in continuous time. We illustrate the most important properties of the individual models and discuss major empirical applications.

We analyze the impact of time series dependence in market microstructure noise on the properties of estimators of the integrated volatility of an asset price based on data sampled at frequencies high enough for that noise to be a dominant consideration. We show that combining two time scales for that purpose will work even when the noise exhibits time series dependence, analyze in that context a refinement of this approach is based on multiple time scales, and compare empirically our different estimators to the standard realized volatility.

The main focus of this work is the analysis of price direction processes at transaction level of different stocks traded at the NYSE. Three different models have been applied to the data, namely the autoregressive conditional ultinomial model, a probit model with latent ARMA-process and different specifications of Markov chain models. The models have been assessed by a description of their strengths and weaknesses in modelling probabilities and by the application of non-nested tests. The implications of the decimalisation at the NYSE on price direction processes were examined in a further application.

This article examines the impact of serial correlation in high frequency returns on the realized variance measure. In particular, it is shown that the realized variance measure yields a biased estimate of the conditional return variance when returns are serially correlated. Using 10 years of FTSE-100 minute by minute data we demonstrate that a careful choice of sampling frequency is crucial in avoiding substantial biases. Moreover, we find that the autocovariance structure (magnitude and rate of decay) of FTSE-100 returns at different sampling frequencies is consistent with that of an ARMA process under temporal aggregation. A simple autocovariance function based method is proposed for choosing the “optimal” sampling frequency, that is, the highest available frequency at which the serial correlation of returns has a negligible impact on the realized variance measure. We find that the logarithmic realized variance series of the FTSE-100 index, constructed using an optimal sampling frequency of 25 minutes, can be modelled as an ARFIMA process. Exogenous variables such as lagged returns and contemporaneous trading volume appear to be highly significant regressors and are able to explain a large portion of the variation in daily realized variance. Dieser Artikel untersucht die Auswirkungen von autokorrelierten Erträgen auf das Maß der realisierten Varianz bei hochfrequenten Daten über die Erträge. Es wird gezeigt, dass die realisierte Varianz ein verzerrter Schätzer für die bedingte Varianz der Erträge bei Vorliegen von Autokorrelation ist. Unter Verwendung eines zehnjährigen Datensatzes von Minutendaten des FTSE-100 wird dargestellt, dass eine sorgfältige Auswahl der Stichprobenfrequenz unabdingbar zur Vermeidung von Verzerrungen ist. Eine einfache Methode zur Bestimmung der optimalen Stichprobenfrequenz, basierend auf der Autokovarianzfunktion, wird vorgeschlagen. Diese ergibt sich als die höchste Frequenz, bei der die vorhandene Autokorrelation noch einen vernachlässigbaren Einfluss auf das Maß der realisierten Varianz hat. Für den betrachteten Datensatz ergibt sich eine optimale Frequenz von 25 Minuten. Unter Verwendung dieser Frequenz können die logarithmierten Erträge des FTSE-100 als ARFIMA Prozess modelliert werden.

In recent years a substantial amount of literature in one way or another deals with liquidity. The interest in it grows beyond the walls of the academia, as the security exchanges recognize the importance of the concept and plan to adopt unique measures of liquidity and publish them in the regular reports. But as in literature, there is still no consensus as what liquidity really means and how it should be measured, or reported, understood or predicted, as the consistent summary of what the liquidity with a meaningful quantitative comparison is practically missing. In the present paper we try to gather and categorise the existent knowledge about liquidity and estimate the measures that can help to gain an idea of the liquidity generated on the transactions level at the stock market.We apply the existent knowledge of liquidity to market microstructure; establish and explain the links between spreads, size of the transactions and price movements and the actual time of the trade. This paper carries out the analysis of the validity of the liquidity measures and transaction based econometric models for five stocks of different trading intensity in order to see how the trading intensity influences the ability to estimate the dynamics of the price formation, information unravelling and evaluate the accuracy of the estimated models.

This paper uses one-minute returns on the TOPIX and S&P500 to examine the efficiency of the Tokyo and New York Stock Exchanges. Our major finding is that Tokyo completes reactions to New York within six minutes, but New York reacts within fourteen minutes. Dividing the sample period into three subperiods, we found that the efficiency has improved and the magnitude of reaction has become larger over the period in both markets. The magnitude of response in New York to a fall in Tokyo is roughly double that of a rise.

This paper proposes a new approach to identifying the effects of monetary policy shocks in an international vector autoregression. Using high-frequency data on the prices of eurodollar contracts, we measure the impact of the surprise component of the FOMC-day Federal Reserve policy decision on financial variables, such as the exchange rate and the foreign interest rate. We show how this information can be used to achieve identification without having to make the usual strong assumption of a recursive ordering.

This paper uses Reuters exchange rate data to investigate thecontributions to the price discovery process by individual banks in theforeign exchange market. We propose multivariate time series models aswell as models in tick time to study the dynamic relations between thequotes of individual banks. We investigate the hypothesis that Germanbanks are price leaders in the deutschmark/dollar market. Our empiricalresults suggest an important but not exclusive role for German banks inthe price discovery process. There is also a group of banks, German andnon-German, that lags behind the market and does not contribute to theprice discovery process. In contrast to Peiers~(1997) we do not findevidence for stronger price leadership of Deutsche bank on days withsuspected Bundesbank interventions in the foreign exchange market.

This paper systematically studies the use of mixed-frequency data sets and suggests that the use of high frequency data in forecasting economic aggregates can improve forecast accuracy. The best way of using this information is to build a single model, for example, an ARMA model with missing observations, that relates data of all frequencies. The implementation of such an approach, however, poses serious practical problems in all but the simplest cases. As a feasible and consistent alternative, we propose a two-stage procedure to obtain pseudo high frequency data and to subsequently use these artificial values as proxies for macroeconomic or financial models. This alternative method yields a sub-optimal forecast in general but avoids the computational problems of a full-blown single model. Our approach differs from classical interpolation since we only use past and current information to get the pseudo series. A proxy, which is constructed by classical interpolation, may fit very well in sample, but it is not useful for out-of-sample forecasts. As applications of linking series generated at different frequencies, we show that the use of monthly proxies of GDP improves the predictability of absolute stock returns and the unemployment rate compared to the use of industrial production as an alternative proxy.