Utilizing open-close returns, close-close returns and volume data, we examine the reaction of the Treasury futures market to the periodically scheduled announcements of prominent U.S. macroeconomic data. Heterogeneous persistence from scheduled news vs. non-scheduled news is revealed. Strong asymmetric effects of scheduled announcements are presented: positive shocks depress volatility on consecutive days, while negative shocks increase volatility. Announcement-day shocks have small persistence, but great impacts on volatility in the short run. Investigation into volume data shows that announcement-day volume has lower persistence than non-announcement-day volume. No statistically significant risk premium manifests on the release dates. Compared with the implied volatility and realized volatility data, we find our model successful in forming both in-sample and out-of-sample multi-step forecasts. Distinctions are made and tested among microstructure theories that differ in predictions of the impact of scheduled macroeconomic news on volatility and volatility persistence. Asymmetric effects between positive and negative shocks from scheduled news call for further exploration of microstructure theory.

Time varying correlations are often estimated with Multivariate Garch models that are linear in squares and cross products of returns. A new class of multivariate models called dynamic conditional correlation (DCC) models is proposed. These have the flexibility of univariate GARCH models coupled with parsimonious parametric models for the correlations. They are not linear but can often be estimated very simply with univariate or two step methods based on the likelihood function. It is shown that they perform well in a variety of situations and give sensible empirical results.

In this paper, we examine the impact of market activity on the percentage bid-ask spreads of S&P 100 index options using transaction data. We propose a new market microstructure theory called a derivative hedge theory, in which option market percentage spreads will be inversely related to the option market maker's ability to hedge his positions in the underlying market, as measured by the liquidity of this underlying market. In a perfect hedge world, spreads arise from the illiquidity of the underlying market, rather than from inventory risk or informed trading in the option market itself.

We estimate three models to investigate various market microstructure theories. In the static model, option spreads are a function of moneyness, time to maturity, option prices, hedge ratios and volatility. The dynamic model includes time between trades or duration and average volume per transaction while the cross-market model adds cross option market activity and spreads in the underlying market.

We find option market volume is not a significant determinant of option market spreads, which challenges the validity of volume as a proxy for liquidity and supports my theory. Option market spreads are positively related to spreads in the underlying market, again supporting our theory. However, option market duration does affect option market spreads, with very slow and very fast option markets both leading to bigger spreads. Only the fast market result would be predicted by asymmetric information theory. Inventory models predict big spreads in slow markets. Neither would be observed if the underlying securities market provided a perfect hedge. We interpret these mixed results to mean that the option market maker is able to only imperfectly hedge his positions in the underlying securities market.

Our result of insignificant option volume casts doubt on the price discovery argument between stock and option markets (Easley, O'Hara, and Srinivas (1997)). Asymmetric information costs in either market are naturally passed to the other market by market maker's hedging and therefore it is unimportant where the informed traders trade.

Recent empirical work has studied point processes of transactions in financial markets and observed clear time dependent patterns in these arrival times. However these studies do not examine the timing of quoted price changes. This paper formulates a bivariate point process to jointly analyze transaction and quote arrivals. In microstructure models, transactions may reveal private information which is then incorporated into new prices. This paper examines the speed of this information flow and the circumstances which govern it. One of the main conclusions are that conditional on past quote times, the impact of trade information is to make quote durations longer when there is more information flow rather than less. This is interpreted as evidence that limit order suppliers become more cautious in the presence of apparent informational trading.

We use Hasbrouck (1991)'s vector autoregressive model for prices and trades to empirically test and assess the role played by the waiting time between consecutive transactions in the process of price formation. We find that as the time duration between transactions decreases, the price impact of trades, the speed of price adjustment to trade related information, and the positive autocorrelation of signed trades, all increase. This suggests that times when markets are most active are times when there is an increased presence of informed traders; we interpret such markets as having reduced liquidity.

This paper aims to bridge the gap between processes where shocks are permanent and those with transitory shocks by formulating a process in which the long run impact of each innovation is time varying and stochastic. Frequent transitory shocks are supplemented by occasional permanent shifts. The stochastic permanent breaks (STOPBREAK) process is based on the premise that a shock is more likely to be permanent if it is large than if it is small. This formulation is motivated by a class of processes that undergo random structural breaks. Consistency and asymptotic normality of quasi maximum likelihood estimates is established and locally best hypothesis tests of the null of a random walk are developed. The model is applied to relative prices of pairs of stocks and significant test statistics result

This paper proposes a new approach to modeling financial transactions data. A new model for discrete valued time series is proposed in the context of generalized linear models. Since the model is specified conditional on both the previous state, as well as the historic distribution, we call the model the Autoregressive Conditional Multinomial (ACM) model. When the data are viewed as a marked point process, the ACD model proposed in Engle and Russell (1998) allows for joint modeling of the price transition probabilities and the arrival times of the transactions. In this marked point process context, the transition probabilities vary continuously through time and are therefore duration dependent. Finally, variations of the model allow for volume and spreads to impact the conditional distribution of price changes. Impulse response studies show the long run price impact of a transaction can be very sensitive to volume but is less sensitive to the spread and transaction rate.

Value at Risk (VaR) has become the standard measure of market risk employed by financial institutions for both internal and regulatory purposes. VaR is defined as the value that a portfolio will lose with a given probability, over a certain time horizon (usually one or ten days). Despite its conceptual simplicity, its measurement is a very challenging statistical problem and none of the methodologies developed so far give satisfactory solutions. Interpreting the VaR as the quantile of future portfolio values conditional on current information, we propose a new approach to quantile estimation which does not require any of the extreme assumptions invoked by existing methodologies (such as normality or i.i.d. returns). The Conditional Autoregressive Value-at-Risk or CAViaR model moves the focus of attention from the distribution of returns directly to the behavior of the quantile. We specify the evolution of the quantile over time using a special type of autoregressive process and use the regression quantile framework introduced by Koenker and Bassett to determine the unknown parameters. Since the objective function is not differentiable, we use a differential evolutionary genetic algorithm for the numerical optimization. Utilizing the criterion that each period the probability of exceeding the VaR must be independent of all the past information, we introduce a new test of model adequacy, the Dynamic Quantile test. Applications to simulated and real data provide empirical support to this methodology and illustrate the ability of these algorithms to adapt to new risk environments.

In this paper we analyze and interpret the quote price dynamics of 100 NYSE stocks with varying average trade frequencies. We specify an error-correction model for the log difference of the bid and the ask price, with the spread acting as the error-correction term, and include as regressors the characteristics of the trades occurring between quote observations, if any. We find that short duration and medium volume trades have the largest impacts on quote prices for all one hundred stocks, and that buyer initiated trades primarily move the ask price while seller initiated trades primarily move the bid price. Trades have a greater impact on quotes in both the short and the long run for the infrequently traded stocks than for the more actively traded stocks. Finally, we find strong evidence that the spread is mean reverting.

In this paper, we develop the theoretical and empirical properties of a new class of multivariate GARCH models capable of estimating large time-varying covariance matrices, Dynamic Conditional Correlation Multivariate GARCH. We show that the problem of multivariate conditional variance estimation can be simplified by estimating univariate GARCH models for each asset, and then, using transformed residuals resulting from the first stage, estimating a conditional correlation estimator. The standard errors for the first stage parameters remain consistent, and only the standard errors for the correlation parameters need be modified. We use the model to estimate the conditional covariance of up to 100 assets using S&P 500 Sector Indices and Dow Jones Industrial Average stocks, and conduct specification tests of the estimator using an industry standard benchmark for volatility models. This new estimator demonstrates very strong performance especially considering ease of implementation of the estimator.