We propose a semiparametric model for the analysis of time series of durations that show autocorrelatins and deterministic patterns. Estimation rests on generalized profile likelihood, which allows for joint estimation of the parametric- an ACD type of model- and nonparametric components, providing consistent and asymptotically normal estimators. It is possible to derive the explicit form for the nonparametric estimator, simplifying estimation to a standard maximum likelihood problem.
This paper studies the dynamic behavior of inflation and unemployment in Spain during the period 1964?1997. In particular, we analyze the implications of high persistence in both unemployment and inflation dynamics for inference regarding the size of Phillips trade-offs and sacrifice ratios in the Spanish economy, in response to a demand shock. To do so we use a Stuctural VAR approach with several identification outlines which give rise to alternative interpretations of the joint unemployment-inflation dynamics. When using a bivariate VAR we cannot reject the existence of a permanent output loss of one-half of one percentage point for each percentage point of permanent disinflation. However, when the VAR is augmented with a third variable, in order to disentangle monetary from non-monetary shocks within the demand class, the evidence favours a lower and marginally permanent trade-off with an output loss of about one-fourth of one percentage point.
General-to-Specific (GETS) modelling has witnessed major advances over the last decade thanks to the automation of multi-path GETS specification search. However, several scholars have argued that the estimation complexity associated with financial models constitutes an obstacle to multi-path GETS modelling in finance. We provide a result with associated methods that overcome many of the problems, and develop a simple but general and flexible algorithm that automates financial multi-path GETS modelling. Starting from a general model where the mean specification can contain autoregressive (AR) terms and explanatory variables, and where the exponential variance specification can include log-ARCH terms, log-GARCH terms, asymmetry terms, Bernoulli jumps and other explanatory variables, the algorithm we propose returns parsimonious mean and variance specifications, and a fat-tailed distribution of the standardised error if normality is rejected. The finite sample properties of the methods and of the algorithm are studied by means of extensive Monte Carlo simulations, and two empirical applications suggest the methods and algorithm are very useful in practice.
This paper proposes estimators of location and size of structural breaks in a, possibly dynamic, nonparametric regression model. The structural breaks can be located at given periods of time and/or they can be explained by the values taken by some regressor, as in threshold models. No previous knowledge of the underlying regression function is required. The paper also studies the case in which several regressors explain the breaks. We derive the rate of convergence and provide Central Limit Theorems for the estimators of the location(s) and size(s). A Monte Carlo experiment illustrates the performance of our estimators in small samples.
We consider impulse response functions to study the impact of both return and volatility on correlation between international equity markets. Using data on US (as the reference country), Canada, UK and France equity indices, empirical evidence shows that without taking into account the effect of return, there is an (asymmetric) effect of volatility on correlation. The volatility seems to have an impact on correlation especially during downturn periods. However, once we introduce the effect of return, the impact of volatility on correlation disappears. These observations suggest that, the relation between volatility and correlation is an association rather than a causality. The strong increase in the correlation is driven by the past of the return and the market direction rather than the volatility.
We constructed a unique data set including price and trade volumes for the Bilbao Stock Exchange (BSE) in the interwar period in order to calculate two alternative market indices (weighted and unweighted). The characteristics of the weekly returns on the market portfolio and trading volumes are analyzed in order to test the existence of various phenomena typical of emerging markets, such as autocorrelation and high persistence of volatility shocks, and other features of advanced markets, such as the risk-return relationship and the relationship between trading volumes and returns. The methodological approach is based on an augmented GARCH-cum-volume model. We find strong evidence in favour of autocorrelation and GARCH effects, no evidence of risk-return relationship, and weak evidence of a contemporaneous impact of trading volumes on returns. These findings are generally in line with the results obtained by recent studies on emerging markets.; Workshop Financial Centres as Competing Clusters, Paris School of Economics, January 30th, 2008
This paper provides an introduction to the problem of modeling randomly spaced
longitudinal data. Although Point Process theory was developed mostly in the sixties
and early seventies, only in the nineties did this field of Probability theory attract the
attention of researchers working in Financial Econometrics. The large increase,
observed since, in the number of different classes of Econometric models for dealing
with financial duration data, has been mostly due to the increased availability of both
trade-by-trade data from equity markets and daily default and rating migration data from
credit markets. This paper provides an overview of the main Econometric models
available in the literature for dealing with what is sometimes called tick data.
Additionally, a synthesis of the basic theory underlying these models is also presented.
Finally, a new theorem dealing with the identifiability of latent intensity factors from
point process data, jointly with a heuristic proof, is introduced.
During the past two decades, innovations protected by patents have played a key role in
business strategies. This fact enhanced studies of the determinants of patents and the impact of
patents on innovation and competitive advantage. Sustaining competitive advantages is as
important as creating them. Patents help sustaining competivite advantages by increasing the
production cost of competitors, by signaling a better quality of products and by serving as
barriers to entry. If patents are rewards for innovation, more R&D should be reflected in more
patents applications but this is not the end of the story. There is empirical evidence showing that
patents through time are becoming easier to get and more valuable to the firm due to increasing
damage awards from infringers. These facts question the constant and static nature of the
relationship between R&D and patents. Furthermore, innovation creates important knowledge
spillovers due to its imperfect appropriability. Our paper investigates these dynamic effects
using U.S. patent data from 1979 to 2000 with alternative model specifications for patent
counts. We introduce a general dynamic count panel data model with dynamic observable and
unobservable spillovers, which encompasses previous models...
Exponential models of autoregressive conditional heteroscedasticity (ARCH) are attractive in empirical analysis because they guarantee the non-negativity of volatility, and because they enable richer autoregressive dynamics. However, the currently available models exhibit stability only for a limited number of conditional densities, and the available estimation and inference methods in the case where the conditional density is unknown hold only under very specific and restrictive assumptions. Here, we provide results and simple methods that readily enables consistent estimation and inference of univariate and multivariate power log-GARCH models under very general and non-restrictive assumptions when the power is fixed, via vector ARMA representations. Additionally, stability conditions are obtained under weak assumptions, and the power log-GARCH model can be viewed as nesting certain classes of stochastic volatility models, including the common ASV(1) specification. Finally, our simulations and empirical applications suggest the model class is very useful in practice.
We compute six different sets of systemic risk measures for a sample of the 20 biggest European and 13 biggest US banks from January 2004 to November 2009. The six measures are based on i) Principal components of the bank’s Credit Default Swaps (CDSs), ii) Interbank interest rate spreads, iii) Structural credit risk models, iv) Collateralized Debt Obligations (CDOs) indexes and their tranches, v) Multivariate densities computed from CDS spreads and vi) Co-Risk measures. We then rank the measures using three different criteria: i) Causality tests, ii) Price discovery tests and iii) their correlation with an index of systemic events. For the European and US markets, the best indicators are the first Principal Component of the single-name CDSs and the LIBOR-OIS or LIBOR-TBILL spreads, respectively, whereas the least reliable indicators are the Co-Risk measures and the systemic spreads extracted from the CDO indexes and their tranches.
This paper develops new methods for determining the cointegration rank in a nonstationary fractionally integrated system, extending univariate optimal methods for testing the degree of integration. We propose a simple Wald test based on the singular value decomposition of the unrestricted estimate of the long run multiplier matrix. When the "strength" of the cointegrating relationship is less than 1/2, the test statistic has a standard asymptotic distribution, like Lagrange Multiplier tests exploiting local properties. We consider the behavior of our test under estimation of short run parameters and local alternatives. We compare our procedure with other cointegration tests based on di erent principles and nd that the new method has better properties in a range of situations by using information on the alternative obtained through a preliminary estimate of the cointegration strength.
We consider regressions of nonstationary fractionally integrated variables dominated by linear time trends. The regression errors are short memory, long memory or even nonstationary, and hence allow for a very flexible cointegration model. In case of simple regressions, least squares estimation gives rise to limiting normal distribucions independently of the order of integration of the regressor, whereas the customary t-statistics diverge. We also investigate the possibility of testing for mean reverting equilibrium deviations by means of a residual-based log-periodogram regression. Asymptotic results become more complicated in the multivariate case.
In this paper we present an equilibrium model of commodity spot (st) and futures (ƒt) prices, with finite elasticity of arbitrage services and convenience yields. By explicitly incorporating and modelling endogenously the convenience yield, our theoretical model is able to capture the existence of backwardation or contango in the long-run spot-futures equilibrium relationship, st = β2ƒt + β3
When the slope of the cointegrating vector β2 > 1(β2 < 1) the market is under long run backwardation (contango). It is the first time in this literature in which the theoretical possibility of finding a cointegrating vector different from the standard β2 = 1 is formally considered.
Independent of the value of β2 this paper shows that the equilibrium model admits an economically meaningful Error Correction Representation, where the linear combination of (st) and (ƒt) characterizing the price discovery process in the framework of Garbade and Silber (1983). coincides exactly with the permanent component of the Gonzalo and Granger (1995) Permanent Transitory decomposition. This linear combination depends on the elasticity of arbitrage seIVices and is determined by the relative liquidity traded in the spot and futures markets. Such outcome not only provides a theoretical justification for this Permanent-Transitory decomposition; but it offers a simple way of detecting which of the two prices is dominant in the price discovery process.
All the results are testable. as can be seen in the application to spot and futures non-ferrous metals prices (Al...
In this paper we present an equilibrium model of commodity spot (st) and futures (ƒt) prices, with finite elasticity of arbitrage services and convenience yields. By explicitly incorporating and modelling endogenously the convenience yield, our theoretical model is able to capture the existence of backwardation or contango in the long-run spot-futures equilibrium relationship, st = β2ƒt + β3 When the slope of the cointegrating vector β2 > 1(β2 < 1) the market is under long run backwardation (contango). It is the first time in this literature in which the theoretical possibility of finding a cointegrating vector different from the standard β2 = 1 is formally considered. Independent of the value of β2 this paper shows that the equilibrium model admits an economically meaningful Error Correction Representation, where the linear combination of (st) and (ƒt) characterizing the price discovery process in the framework of Garbade and Silber (1983). coincides exactly with the permanent component of the Gonzalo and Granger (1995) Permanent Transitory decomposition. This linear combination depends on the elasticity of arbitrage seIVices and is determined by the relative liquidity traded in the spot and futures markets. Such outcome not only provides a theoretical justification for this Permanent-Transitory decomposition; but it offers a simple way of detecting which of the two prices is dominant in the price discovery process. All the results are testable. as can be seen in the application to spot and futures non-ferrous metals prices (Al...
The paper examines a Lagrange Multiplier type test for the constancy of the parameter in general models with dependent data without imposing any artificial choice of the possible location of the break. In order to prove the asymptotic behaviour of the test, we extend a strong approximation result for partial sums of a sequence of random variables. We also present a Monte-Carlo experiment to examine the finite sample performance of the test and how it compares with tests which assume some knowledge of the possible location of the break.; The rst author gratefully acknowledges the research support by a Catedra of Excellence by the Bank of Santander.
Dynamic interactions among stock return, Research and Development (R&D) expenses, patent applications based on R&D investment, and the propensity to patent are studied in this work for a panel of firms from the United States. The panel includes technologically similar firms, neck-to-neck, mostly from the drugs product-market sector. Firms’ propensity to patent is modeled by a dynamic latent-factor patent count data model that separates patented and non patented R&D. Patent innovation leader and follower firms are identified according to their knowledge stock. Significant and positive dynamic spillover effects are obtained among patent application leaders and followers. We observe that neck-to-neck firms in patent innovation activity produce an inverted-U relationship between market competition and innovation. Furthermore, firms’ propensity to patent is positively correlated with market competition and there is a positive feedback in both directions. Increasing the degree of competition in the market enhances innovation and patent applications, in order to help firms to appropriate part of the benefits of
their R&D investments. On the other hand, firms by increasing their patent applications defend themselves from competitors, trying to improve their market share. However...
The parameters of popular multivariate GARCH (MGARCH) models are restricted so that their estimation is feasible in large systems and covariance stationarity and positive definiteness of conditional covariance matrices are guaranteed. These restrictions limit the dynamics that the models can represent, assuming, for example, that volatilities evolve in an univariate fashion, not being related neither among them nor with the correlations. This paper updates previous surveyson parametric MGARCH models focusing on their limitations to represent the dynamics observed in real systems of financial returns. The conclusions are illustrated using simulated data and a five-dimensional system of exchange rate returns.; The first author was supported by grants of 0969/13-3 CAPES, Coordination
of Improvement of Higher Education Personnel. The second author acknowledges financial support
from CAPES, grant 10600/13-2, São Paulo Research Foundation (FAPESP), grant 2013/00506-1,
and Laboratory EPIFISMA. Financial support from ECO2012-32401 project by the Spanish
Government is gratefully acknowledged by the third author.
Glutathione transferases (GSTs) are dimeric enzymes containing one active-site per monomer. The omega-class GSTs (hGSTO1-1 and hGSTO2-2 in humans) are homodimeric and carry out a range of reactions including the glutathione-dependant reduction of a range of compounds and the reduction of S-(phenacyl)glutathiones to acetophenones. Both types of reaction result in the formation of a mixed-disulfide of the enzyme with glutathione through the catalytic cysteine (C32). Recycling of the enzyme utilizes a second glutathione molecule and results in oxidized glutathione (GSSG) release. The crystal structure of an active-site mutant (C32A) of the hGSTO1-1 isozyme in complex with GSSG provides a snapshot of the enzyme in the process of regeneration. GSSG occupies both the G (GSH-binding) and H (hydrophobic-binding) sites and causes re-arrangement of some H-site residues. In the same structure we demonstrate the existence of a novel "ligandin" binding site deep within in the dimer interface of this enzyme, containing S-(4-nitrophenacyl)glutathione, an isozyme-specific substrate for hGSTO1-1. The ligandin site, conserved in Omega class GSTs from a range of species, is hydrophobic in nature and may represent the binding location for tocopherol esters that are uncompetitive hGSTO1-1 inhibitors.; This work was supported by National Health and Medical Research Council Project Grant 366731. AJO is supported by an Australian Research Council
Future Fellowship FT0990287.
This paper studies the dynamic interactions and the spillovers that exist among patent application intensity, secret innovation intensity and stock returns of a well-defined technological cluster of firms. We study the differential behavior when there is an Innovation Leader (IL) and the rest of the firms are Innovation Followers (IFs). The leader and the followers of the technological cluster are defined according to their patent innovation activity (stock of knowledge). We use data on stock returns and patent applications of a panel of technologically related firms of the United States (US) economy over the period 1979 to 2000. Most firms of the technological cluster are from the pharmaceutical-products industry. Interaction effects and spillovers are quantified by applying several Panel Vector Autoregressive (PVAR) market value models. Impulse Response Functions (IRFs) and dynamic interaction multipliers of the PVAR models are estimated. Secret patent innovations are estimated by using a recent Poisson-type patent count data model, which includes a set of dynamic latent variables. We show that firms’ stock returns, observable patent intensities and secret patent intensities have significant dynamic interaction effects for technologically related firms. The predictive absorptive capacity of the IL is the highest and this type of absorptive capacity is positively correlated with good firm performance measures. The innovation spillover effects that exist among firms...
Using the conventional VAR identification approach, Cochrane (1994) finds that substantial amounts of variation in GDP growth and stock returns are due to transitory shocks. Following the common trend decomposition of King, et al. (1991), we show that Cochrane's results depend on the assumption of weak exogeneity of one of the variables with respect to the cointegration vector. When this assumption holds both approaches coincide. If not, the shocks Cochrane called transitory are not totally transitory. In this case, the conventional VAR approach with the assumption of the weak exogeneity may overstate the magnitude of transitory shocks and understate that of permanent shocks. We find that the permanent components of GDP and stock prices are much larger than those estimates of Cochrane, although substantial (but much smaller than in Cochrane (1994)) variations in GDP growth and stock returns are attributed to transitory shocks.