Testing the random walk hypothesis with variance ratio statistics

Mestrado em Finanças

A method to quantify a descriptor's illumination variance

This paper presents a new metric, which we call the lighting variance ratio, for quantifying descriptors in terms of their variance to illumination changes. In many applications it is desirable to have descriptors that are robust to changes in illumination, especially in outdoor environments. The lighting variance ratio is useful for comparing descriptors and determining if a descriptor is lighting invariant enough for a given environment. The metric is analysed across a number of datasets, cameras and descriptors. The results show that the upright SIFT descriptor is typically the most lighting invariant descriptor.

Testing Mean-Variance Efficiency in CAPM with Possibly Non-Gaussian Errors : An Exact Simulation-Based Approach

In this paper we propose exact likelihood-based mean-variance efficiency tests of the market portfolio in the context of Capital Asset Pricing Model (CAPM), allowing for a wide class of error distributions which include normality as a special case. These tests are developed in the frame-work of multivariate linear regressions (MLR). It is well known however that despite their simple statistical structure, standard asymptotically justified MLR-based tests are unreliable. In financial econometrics, exact tests have been proposed for a few specific hypotheses [Jobson and Korkie (Journal of Financial Economics, 1982), MacKinlay (Journal of Financial Economics, 1987), Gib-bons, Ross and Shanken (Econometrica, 1989), Zhou (Journal of Finance 1993)], most of which depend on normality. For the gaussian model, our tests correspond to Gibbons, Ross and Shanken’s mean-variance efficiency tests. In non-gaussian contexts, we reconsider mean-variance efficiency tests allowing for multivariate Student-t and gaussian mixture errors. Our framework allows to cast more evidence on whether the normality assumption is too restrictive when testing the CAPM. We also propose exact multivariate diagnostic checks (including tests for multivariate GARCH and mul-tivariate generalization of the well known variance ratio tests) and goodness of fit tests as well as a set estimate for the intervening nuisance parameters. Our results [over five-year subperiods] show the following: (i) multivariate normality is rejected in most subperiods, (ii) residual checks reveal no significant departures from the multivariate i.i.d. assumption, and (iii) mean-variance efficiency tests of the market portfolio is not rejected as frequently once it is allowed for the possibility of non-normal errors.

Informational efficiency of the European Union emission allowance market

This thesis studies the informational efficiency of the European Union emission allowance (EUA) market. In an efficient market, the market price is unpredictable and profits above average are impossible in the long run. The main research problem is does the EUA price follow a random walk. The method is an econometric analysis of the price series, which includes an autocorrelation coefficient test and a variance ratio test. The results reveal that the price series is autocorrelated and therefore a nonrandom walk. In order to find out the extent of predictability, the price series is modelled with an autoregressive model. The conclusion is that the EUA price is autocorrelated only to a small degree and that the predictability cannot be used to make extra profits. The EUA market is therefore considered informationally efficient, although the price series does not fulfill the requirements of a random walk. A market review supports the conclusion, but it is clear that the maturing of the market is still in process.

Robust and High-resolution Voiced/Unvoiced Classification in Noisy Speech Using A Signal Smoothness Criterion

We propose a novel technique for robust voiced/unvoiced segment detection in noisy speech, based on local polynomial regression. The local polynomial model is well-suited for voiced segments in speech. The unvoiced segments are noise-like and do not exhibit any smooth structure. This property of smoothness is used for devising a new metric called the variance ratio metric, which, after thresholding, indicates the voiced/unvoiced boundaries with 75% accuracy for 0dB global signal-to-noise ratio (SNR). A novelty of our algorithm is that it processes the signal continuously, sample-by-sample rather than frame-by-frame. Simulation results on TIMIT speech database (downsampled to 8kHz) for various SNRs are presented to illustrate the performance of the new algorithm. Results indicate that the algorithm is robust even in high noise levels.

Conditional Moments in Asset Pricing

A better understanding of stock price changes is important in guiding many economic activities. Since prices often do not change without good reasons, searching for related explanatory variables has involved many enthusiasts. This book seeks answers from prices per se by relating price changes to their conditional moments. This is based on the belief that prices are the products of a complex psychological and economic process and their conditional moments derive ultimately from these psychological and economic shocks. Utilizing information about conditional moments hence makes it an attractive alternative to using other selective financial variables in explaining price changes. The first paper examines the relation between the conditional mean and the conditional variance using information about moments in three types of conditional distributions; it finds that the significance of the estimated mean and variance ratio can be affected by the assumed distributions and the time variations in skewness. The second paper decomposes the conditional industry volatility into a concurrent market component and an industry specific component; it finds that market volatility is on average responsible for a rather small share of total industry volatility — 6 to 9 percent in UK and 2 to 3 percent in Germany. The third paper looks at the heteroskedasticity in stock returns through an ARCH process supplemented with a set of conditioning information variables; it finds that the heteroskedasticity in stock returns allows for several forms of heteroskedasticity that include deterministic changes in variances due to seasonal factors, random adjustments in variances due to market and macro factors, and ARCH processes with past information. The fourth paper examines the role of higher moments — especially skewness and kurtosis — in determining the expected returns; it finds that total skewness and total kurtosis are more relevant non-beta risk measures and that they are costly to be diversified due either to the possible eliminations of their desirable parts or to the unsustainability of diversification strategies based on them.

Are component positioning and prosthesis size associated with hip resurfacing failure?

BACKGROUND: Recent studies suggest that there is a learning curve for metal-on-metal hip resurfacing. The purpose of this study was to assess whether implant positioning changed with surgeon experience and whether positioning and component sizing were associated with implant longevity. METHODS: We evaluated the first 361 consecutive hip resurfacings performed by a single surgeon, which had a mean follow-up of 59 months (range, 28 to 87 months). Pre and post-operative radiographs were assessed to determine the inclination of the acetabular component, as well as the sagittal and coronal femoral stem-neck angles. Changes in the precision of component placement were determined by assessing changes in the standard deviation of each measurement using variance ratio and linear regression analysis. Additionally, the cup and stem-shaft angles as well as component sizes were compared between the 31 hips that failed over the follow-up period and the surviving components to assess for any differences that might have been associated with an increased risk for failure. RESULTS: Surgeon experience was correlated with improved precision of the antero-posterior and lateral positioning of the femoral component. However, femoral and acetabular radiographic implant positioning angles were not different between the surviving hips and failures. The failures had smaller mean femoral component diameters as compared to the non-failure group (44 versus 47 millimeters). CONCLUSIONS: These results suggest that there may be differences in implant positioning in early versus late learning curve procedures, but that in the absence of recognized risk factors such as intra-operative notching of the femoral neck and cup inclination in excess of 50 degrees, component positioning does not appear to be associated with failure. Nevertheless, surgeons should exercise caution in operating patients with small femoral necks, especially when they are early in the learning curve.

Margin requirements and volatility : evidence from Canadian stocks

Margin policy is used by regulators for the purpose of inhibiting exceSSIve volatility and stabilizing the stock market in the long run. The effect of this policy on the stock market is widely tested empirically. However, most prior studies are limited in the sense that they investigate the margin requirement for the overall stock market rather than for individual stocks, and the time periods examined are confined to the pre-1974 period as no change in the margin requirement occurred post-1974 in the U.S. This thesis intends to address the above limitations by providing a direct examination of the effect of margin requirement on return, volume, and volatility of individual companies and by using more recent data in the Canadian stock market. Using the methodologies of variance ratio test and event study with conditional volatility (EGARCH) model, we find no convincing evidence that change in margin requirement affects subsequent stock return volatility. We also find similar results for returns and trading volume. These empirical findings lead us to conclude that the use of margin policy by regulators fails to achieve the goal of inhibiting speculating activities and stabilizing volatility.

Finite-Sample Diagnostics for Multivariate Regressions with Applications to Linear Asset Pricing Models

In this paper, we propose several finite-sample specification tests for multivariate linear regressions (MLR) with applications to asset pricing models. We focus on departures from the assumption of i.i.d. errors assumption, at univariate and multivariate levels, with Gaussian and non-Gaussian (including Student t) errors. The univariate tests studied extend existing exact procedures by allowing for unspecified parameters in the error distributions (e.g., the degrees of freedom in the case of the Student t distribution). The multivariate tests are based on properly standardized multivariate residuals to ensure invariance to MLR coefficients and error covariances. We consider tests for serial correlation, tests for multivariate GARCH and sign-type tests against general dependencies and asymmetries. The procedures proposed provide exact versions of those applied in Shanken (1990) which consist in combining univariate specification tests. Specifically, we combine tests across equations using the MC test procedure to avoid Bonferroni-type bounds. Since non-Gaussian based tests are not pivotal, we apply the “maximized MC” (MMC) test method [Dufour (2002)], where the MC p-value for the tested hypothesis (which depends on nuisance parameters) is maximized (with respect to these nuisance parameters) to control the test’s significance level. The tests proposed are applied to an asset pricing model with observable risk-free rates, using monthly returns on New York Stock Exchange (NYSE) portfolios over five-year subperiods from 1926-1995. Our empirical results reveal the following. Whereas univariate exact tests indicate significant serial correlation, asymmetries and GARCH in some equations, such effects are much less prevalent once error cross-equation covariances are accounted for. In addition, significant departures from the i.i.d. hypothesis are less evident once we allow for non-Gaussian errors.

Validation of three new measure-correlate-predict models for the long-term prospection of the wind resource

The estimation of the long-term wind resource at a prospective site based on a relatively short on-site measurement campaign is an indispensable task in the development of a commercial wind farm. The typical industry approach is based on the measure-correlate-predict �MCP� method where a relational model between the site wind velocity data and the data obtained from a suitable reference site is built from concurrent records. In a subsequent step, a long-term prediction for the prospective site is obtained from a combination of the relational model and the historic reference data. In the present paper, a systematic study is presented where three new MCP models, together with two published reference models �a simple linear regression and the variance ratio method�, have been evaluated based on concurrent synthetic wind speed time series for two sites, simulating the prospective and the reference site. The synthetic method has the advantage of generating time series with the desired statistical properties, including Weibull scale and shape factors, required to evaluate the five methods under all plausible conditions. In this work, first a systematic discussion of the statistical fundamentals behind MCP methods is provided and three new models, one based on a nonlinear regression and two �termed kernel methods� derived from the use of conditional probability density functions, are proposed. All models are evaluated by using five metrics under a wide range of values of the correlation coefficient, the Weibull scale, and the Weibull shape factor. Only one of all models, a kernel method based on bivariate Weibull probability functions, is capable of accurately predicting all performance metrics studied.

Assessment of inbreeding depression in a Guzerat dairy herd: Effects of individual increase in inbreeding coefficients on production and reproduction

Influences of inbreeding on daily milk yield (DMY), age at first calving (AFC), and calving intervals (CI) were determined on a highly inbred zebu dairy subpopulation of the Guzerat breed. Variance components were estimated using animal models in single-trait analyses. Two approaches were employed to estimate inbreeding depression: using individual increase in inbreeding coefficients or using inbreeding coefficients as possible covariates included in the statistical models. The pedigree file included 9,915 animals, of which 9,055 were inbred, with an average inbreeding coefficient of 15.2%. The maximum inbreeding coefficient observed was 49.45%, and the average inbreeding for the females still in the herd during the analysis was 26.42%. Heritability estimates were 0.27 for DMY and 0.38 for AFC. The genetic variance ratio estimated with the random regression model for CI ranged around 0.10. Increased inbreeding caused poorer performance in DMY, AFC, and CI. However, some of the cows with the highest milk yield were among the highly inbred animals in this subpopulation. Individual increase in inbreeding used as a covariate in the statistical models accounted for inbreeding depression while avoiding overestimation that may result when fitting inbreeding coefficients.

Efficiency, cointegration and contagion in equity markets : evidence from China, Japan and South Korea

This paper empirically examines whether three East Asian stock markets, namely, those of China, Japan and South Korea, are individually and/or jointly efficient, and whether contagion exists between the cointegrated markets. While individual market efficiency is examined through testing for the random walk hypothesis, joint market efficiency is examined through testing for cointegration and contagion. The present study finds that the hypothesis of individual market efficiency is strongly rejected for the Chinese stock market, but not for the Japanese and the South Korean stock markets. However, when testing for cointegration, market efficiency is strongly rejected for all these markets. We take a simple case of contagion and find that although there is a long-term relationship among the three markets, the contagion hypothesis cannot be rejected only between Japanese and South Korean stock markets, indicating short-run portfolio diversification benefits from these two markets.

External shocks and tourist arrivals to Cambodia

This article investigates how external shocks affect tourist arrivals to Cambodia. The study relies on the random walk approach to test whether the shocks to tourist arrivals are temporary or persistent in nature. To facilitate the empirical investigation, the study applies and compares the results from different unit root tests and variance-ratio tests to the monthly tourists' arrival data from 1994 to December 2012. Both tests provide evidence of random walk hypothesis, implying that shocks to tourists' arrival to Cambodia have a persistent effect requiring short- to medium-term policies to combat the vulnerability due to those shocks. Public-private coordinated policies could reduce the impact, as we found that once the crisis (external shocks) is minimized, the magnitude of the shocks decays slowly.

Interday and intraday volatility: additional evidence from the Shanghai Stock Exchange

After examining both the interday and intraday return volatility of the Shanghai Composite Stock Index, it was found that the open-to-open return variance is consistently greater than the close-to-close variance. Examining the volatility of interday returns and variance ratio tests with five-minute intervals reveals an L-shaped pattern, or more precisely, two L-shaped patterns, starting with a small hump during both the morning and the afternoon sessions, with the morning session having a much higher interday volatility than the afternoon session. This L -shaped interday volatility is supported by the similarly shaped intraday volatility pattern. This result suggests that the high volatility of intraday returns for the market open is not entirely due to the trading mechanisms (call auction in the market opening) but also due to both the accumulated overnight information and the trading halt effect. The five-minute breaks after the auction and blind auction procedures are the two major driving forces which exaggerate the high intraday volatility observed at the market open.