An autoregressive conditional filtering process to remove intraday seasonal volatility and its application to testing the noisy rational expectations model

We develop a new autoregressive conditional process to capture both the changes and the persistency of the intraday seasonal (U-shape) pattern of volatility in essay 1. Unlike other procedures, this approach allows for the intraday volatility pattern to change over time without the filtering process injecting a spurious pattern of noise into the filtered series. We show that prior deterministic filtering procedures are special cases of the autoregressive conditional filtering process presented here. Lagrange multiplier tests prove that the stochastic seasonal variance component is statistically significant. Specification tests using the correlogram and cross-spectral analyses prove the reliability of the autoregressive conditional filtering process. In essay 2 we develop a new methodology to decompose return variance in order to examine the informativeness embedded in the return series. The variance is decomposed into the information arrival component and the noise factor component. This decomposition methodology differs from previous studies in that both the informational variance and the noise variance are time-varying. Furthermore, the covariance of the informational component and the noisy component is no longer restricted to be zero. The resultant measure of price informativeness is defined as the informational variance divided by the total variance of the returns. The noisy rational expectations model predicts that uninformed traders react to price changes more than informed traders, since uninformed traders cannot distinguish between price changes caused by information arrivals and price changes caused by noise. This hypothesis is tested in essay 3 using intraday data with the intraday seasonal volatility component removed, as based on the procedure in the first essay. The resultant seasonally adjusted variance series is decomposed into components caused by unexpected information arrivals and by noise in order to examine informativeness.

An Autoregressive Conditional Filtering Process to remove Intraday Seasonal Volatility and its Application to Testing the Noisy Rational Expectations Model

We develop a new autoregressive conditional process to capture both the changes and the persistency of the intraday seasonal (U-shape) pattern of volatility in essay 1. Unlike other procedures, this approach allows for the intraday volatility pattern to change over time without the filtering process injecting a spurious pattern of noise into the filtered series. We show that prior deterministic filtering procedures are special cases of the autoregressive conditional filtering process presented here. Lagrange multiplier tests prove that the stochastic seasonal variance component is statistically significant. Specification tests using the correlogram and cross-spectral analyses prove the reliability of the autoregressive conditional filtering process. In essay 2 we develop a new methodology to decompose return variance in order to examine the informativeness embedded in the return series. The variance is decomposed into the information arrival component and the noise factor component. This decomposition methodology differs from previous studies in that both the informational variance and the noise variance are time-varying. Furthermore, the covariance of the informational component and the noisy component is no longer restricted to be zero. The resultant measure of price informativeness is defined as the informational variance divided by the total variance of the returns. The noisy rational expectations model predicts that uninformed traders react to price changes more than informed traders, since uninformed traders cannot distinguish between price changes caused by information arrivals and price changes caused by noise. This hypothesis is tested in essay 3 using intraday data with the intraday seasonal volatility component removed, as based on the procedure in the first essay. The resultant seasonally adjusted variance series is decomposed into components caused by unexpected information arrivals and by noise in order to examine informativeness.

Three essays on the microstructure of exchange traded funds

Exchange traded funds (ETFs) have increased significantly in popularity since they were first introduced in 1993. However, there is still much that is unknown about ETFs in the extant literature. This dissertation attempts to fill gaps in the ETF literature by using three related essays. In these three essays, we compare ETFs to closed ended mutual funds (CEFs) by decomposing the bid-ask spread into its three components; we look at the intraday shape of ETFs and compare it to the intraday shape of equities as well as examine the co-integration factor between ETFs on the London Stock Exchange and the New York Stock Exchange; we also examine the differences between leveraged ETFs and unleveraged ETFs by analyzing the impact of liquidity and volatility. These three essays are presented in Chapters 1, 2, and 3, respectively. ^ Chapter one uses the Huang and Stoll (1997) model to decompose the bid-ask spread in CEFs and ETFs for two distinct periods—a normal and a volatile period. We show a higher adverse selection component for CEFs than for ETFs without regard to volatility. However, both ETFs and CEFs increased in magnitude of the adverse selection component in the period of high volatility. Chapter two uses a mix of the Werner and Kleidon (1993) and the Hupperets and Menkveld (2002) methods to get the intraday shape of ETFs and analyze co-integration between London and New York trading. We find two different shapes for New York and London ETFs. There also appears to be evidence of co-integration in the overlapping two-hour trading period but not over the entire trading day for the two locations. The third chapter discusses the new class of ETFs called leveraged ETFs. We examine the liquidity and depth differences between unleveraged and leveraged ETFs at the aggregate level and when the leveraged ETFs are classified by the leveraged multiples of -3, -2, -1, 2, and 3, both for a normal and a volatile period. We find distinct differences between leveraged and unleveraged ETFs at the aggregate level, with leveraged ETFs having larger spreads than unleveraged ETFs. Furthermore, while both leveraged and unleveraged ETFs have larger spreads in high volatility, for the leveraged ETFs the change in magnitude is significantly larger than for the unleveraged ETFs. Among the multiples, the -2 leveraged ETF is the most pronounced in its liquidity characteristics, more so in volatile times. ^