Zhou, Tu. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (leaves 61-70). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.0 / Chapter 1.1 --- Motivation and Objective --- p.0 / Chapter 1.2 --- Major Contribution --- p.2 / Chapter 1.3 --- Thesis Organization --- p.4 / Chapter 2 --- Measurement of Relationship between financial time series --- p.5 / Chapter ´ب2.1 --- Linear Correlation --- p.5 / Chapter 2.1.1 --- Pearson Correlation Coefficient --- p.6 / Chapter 2.1.2 --- Rank Correlation --- p.6 / Chapter 2.2 --- Mutual Information --- p.7 / Chapter 2.2.1 --- Approaches of Mutual Information Estimation --- p.10 / Chapter 2.3 --- Copula --- p.12 / Chapter 2.4 --- Analysis from Experimental Data --- p.14 / Chapter 2.4.1 --- Experiment 1: Nonlinearity --- p.14 / Chapter 2.4.2 --- Experiment 2: Sensitivity of Outliers --- p.16 / Chapter 2.4.3 --- Experiment 3: Transformation Invariance --- p.20 / Chapter 2.5 --- Chapter Summary --- p.23 / Chapter 3 --- Clustered Dynamic Conditional Correlation Model --- p.26 / Chapter 3.1 --- Background Review --- p.26 / Chapter 3.1.1 --- GARCH Model --- p.26 / Chapter 3.1.2 --- Multivariate GARCH model --- p.29 / Chapter 3.2 --- DCC Multivariate GARCH Models --- p.31 / Chapter 3.2.1 --- DCC GARCH Model --- p.31 / Chapter 3.2.2 --- Generalized DCC GARCH Model --- p.32 / Chapter 3.2.3 --- Block-DCC GARCH Model --- p.32 / Chapter 3.3 --- Clustered DCC GARCH Model --- p.34 / Chapter 3.3.1 --- Minimum Distance Estimation (MDE) --- p.36 / Chapter 3.3.2 --- Clustered DCC (CDCC) based on MDE --- p.37 / Chapter 3.4 --- Clustering Method Selection --- p.40 / Chapter 3.5 --- Model Estimation and Testing Method --- p.42 / Chapter 3.5.1 --- Maximum Likelihood Estimation --- p.42 / Chapter 3.5.2 --- Box-Pierce Statistic Test --- p.44 / Chapter 3.6 --- Chapter Summary --- p.44 / Chapter 4 --- Experimental Result and Applications on CDCC --- p.46 / Chapter 4.1 --- Model Comparison and Analysis --- p.46 / Chapter 4.2 --- Portfolio Selection Application --- p.50 / Chapter 4.3 --- Value at Risk Application --- p.52 / Chapter 4.4 --- Chapter Summary --- p.55 / Chapter 5 --- Conclusion --- p.57 / Bibliography --- p.61

Abstract not available

This thesis examines the relations among emotions and musical features and their changes with time, based on the assertion that there exist underlying, culturally specific, quantifiable rules which govern these relations. I designed, programmed and tested a computer controlled Two-Dimensional Emotion Space (2DES) which administered and controlled all aspects of the experimental work. The 2DES instrument consisted of two bipolar emotional response (ER) dimensions: valence (happiness-sadness) and arousal (activeness-sleepiness). The instrument had a test-retest reliability exceeding 0.83 (p &gt 0.01, N = 28) when words and pictures of facial expressions were used as the test stimuli. Construct validity was quantified (r &lt 0.84, p &gt 0.01). The 2DES was developed to collect continuous responses to recordings of four movements of music (N = 67) chosen to elicit responses in all quadrants of the 2DES: &quotMorning&quot from Peer Gynt, Adagio from Rodrigo???s Concierto de Aranjuez (Aranjuez), Dvorak???s Slavonic Dance Op 42, No. 1 and Pizzicato Polka by Strauss. Test-retest reliability was 0.74 (p &gt 0.001, N = 14). Five salient and objectively quantifiable features of the musical signal (MFs) were scaled and used for time series analysis of the stimuli: melodic pitch, tempo, loudness, frequency spectrum centroid (timbral sharpness) and texture (number of different instruments playing). A quantitative analysis consisted of: (1) first order differencing to remove trends, (2) determination of suitable, lagged MFs to keep as regressors via stepwise regression, and (3) regression of each ER onto selected MFs with first order autoregressive adjustment for serial correlation. Regression coefficients indicated that first order differenced (???) loudness and ???tempo had the largest correlations with ???arousal across all pieces, and ???melodic pitch correlated with ???valence for Aranjuez (p &gt 0.01 for all coefficients). The models were able to explain up to 73% of mean response variance. Additional variation was explained qualitatively as being due to interruptions, interactions and collinearity: The minor key and dissonances in a tonal context moved valence toward the negative direction; Short duration and perfect cadences moved valence in the positive direction. The 2DES measure and serial correlation adjusted regression models were, together, shown to be powerful tools for understanding relations among musical features and emotional response.

Amar G. Bose. / "May 15, 1956." "This report is based on a thesis submitted to the Department of Electrical Engineering, M.I.T., May 14, 1956, in partial fulfillment of the requirements for the degree of Doctor of Science." / Bibliography: p. 58. / Army Signal Corps Contract DA36-039-sc-64637 Dept. of the Army Task No. 3-99-06-108 Project No. 3-99-00-100

John P. Costas. / "May 16, 1950." / Bibliography: p. 7. / Army Signal Corps Contract No. W36-039-sc-32037 Project No. 102B Dept. of the Army Project No. 3-99-10-022

Time series analysis of Synthetic Aperture Radar Interferometry (InSAR) data has become an important scientific tool for monitoring and measuring the displacement of Earth’s surface due to a wide range of phenomena, including earthquakes, volcanoes,landslides, changes in ground water levels, and wetlands. Time series analysis is a product of interferometric phase measurements, which become ambiguous when the observed motion is larger than half of the radar wavelength. Thus, phase observations must first be unwrapped in order to obtain physically meaningful results. Persistent Scatterer Interferometry (PSI), Stanford Method for Persistent Scatterers (StaMPS), Short Baselines Interferometry (SBAS) and Small Temporal Baseline Subset (STBAS)algorithms solve for this ambiguity using a series of spatio-temporal unwrapping algorithms and filters. In this dissertation, I improve upon current phase unwrapping algorithms, and apply the PSI method to study subsidence in Mexico City. PSI was used to obtain unwrapped deformation rates in Mexico City (Chapter 3),where ground water withdrawal in excess of natural recharge causes subsurface, clay-rich sediments to compact. This study is based on 23 satellite SAR scenes acquired between January 2004 and July 2006. Time series analysis of the data reveals a maximum line-of-sight subsidence rate of 300mm/yr at a high enough resolution that individual subsidence rates for large buildings can be determined. Differential motion and related structural damage along an elevated metro rail was evident from the results. Comparison of PSI subsidence rates with data from permanent GPS stations indicate root mean square(RMS) agreement of 6.9 mm/yr, about the level expected based on joint data uncertainty.The Mexico City results suggest negligible recharge, implying continuing degradation and loss of the aquifer in the third largest metropolitan area in the world. Chapters 4 and 5 illustrate the link between time series analysis and three-dimensional (3-D) phase unwrapping. Chapter 4 focuses on the unwrapping path.Unwrapping algorithms can be divided into two groups, path-dependent and path-independent algorithms. Path-dependent algorithms use local unwrapping functions applied pixel-by-pixel to the dataset. In contrast, path-independent algorithms use global optimization methods such as least squares, and return a unique solution. However, when aliasing and noise are present, path-independent algorithms can underestimate the signal in some areas due to global fitting criteria. Path-dependent algorithms do not underestimate the signal, but, as the name implies, the unwrapping path can affect the result. Comparison between existing path algorithms and a newly developed algorithm based on Fisher information theory was conducted. Results indicate that Fisher information theory does indeed produce lower misfit results for most tested cases. Chapter 5 presents a new time series analysis method based on 3-D unwrapping of SAR data using extended Kalman filters. Existing methods for time series generation using InSAR data employ special filters to combine two-dimensional (2-D) spatial unwrapping with one-dimensional (1-D) temporal unwrapping results. The new method,however, combines observations in azimuth, range and time for repeat pass interferometry. Due to the pixel-by-pixel characteristic of the filter, the unwrapping path is selected based on a quality map. This unwrapping algorithm is the first application of extended Kalman filters to the 3-D unwrapping problem. Time series analyses of InSAR data are used in a variety of applications with different characteristics. Consequently, it is difficult to develop a single algorithm that can provide optimal results in all cases, given that different algorithms possess a unique set of strengths and weaknesses. Nonetheless, filter-based unwrapping algorithms such as the one presented in this dissertation have the capability of joining multiple observations into a uniform solution, which is becoming an important feature with continuously growing datasets.

Thesis (M. Phil.)--University of Hong Kong, 2006. / Also available in print.

Ph.D. / Electrical and Computer Engineering / Numerous applications require either the estimation or prediction of a noisy time-series. Examples include speech enhancement, economic forecasting, and geophysical modeling. A noisy time-series can be described in terms of a probabilistic model, which accounts for both the deterministic and stochastic components of the dynamics. Such a model can be used with a Kalman filter (or extended Kalman filter) to estimate and predict the time-series from noisy measurements. When the model is unknown, it must be estimated as well; dual estimation refers to the problem of estimating both the time-series, and its underlying probabilistic model, from noisy data. The majority of dual estimation techniques in the literature are for signals described by linear models, and many are restricted to off-line application domains. Using a probabilistic approach to dual estimation, this work unifies many of the approaches in the literature within a common theoretical and algorithmic framework, and extends their capabilities to include sequential dual estimation of both linear and nonlinear signals. The dual Kalman filtering method is developed as a method for minimizing a variety of dual estimation cost functions, and is shown to be an effective general method for estimating the signal, model parameters, and noise variances in both on-line and off-line environments.

In this dissertation, we propose a set of computationally efficient methods based on approximating/representing nonlinear processes by linear ones, so-called linearization. Firstly, a linearization method is introduced for estimating the multiple frequencies in sinusoidal processes. It utilizes a regularized autoregressive (AR) approximation, which can be regarded as a "large p - small n" approach in a time series context. An appealing property of regularized AR is that it avoids a model selection step and allows for an efficient updating of the frequency estimates whenever new observations are obtained. The theoretical analysis shows that the regularized AR frequency estimates are consistent and asymptotically normally distributed. Secondly, a sieve bootstrap scheme is proposed using the linear representation of generalized autoregressive conditional heteroscedastic (GARCH) models to construct prediction intervals (PIs) for the returns and volatilities. Our method is simple, fast and distribution-free, while providing sharp and well-calibrated PIs. A similar linear bootstrap scheme can also be used for diagnostic testing. Thirdly, we introduce a robust lagrange multiplier (LM) test, which utilizes either the bootstrap or permutation procedure to obtain critical values, for detecting GARCH effects. We justify that both bootstrap and permutation LM tests are consistent. Intensive numerical studies indicate that the proposed resampling algorithms significantly improve the size and power of the LM test in both skewed and heavy-tailed processes. Moreover, fourthly, we introduce a nonparametric trend test in the presence of GARCH effects (NT-GARCH) based on heteroscedastic ANOVA. Our empirical evidence show that NT-GARCH can effectively detect non-monotonic trends under GARCH, especially in the presence of irregular seasonal components. We suggest to apply the bootstrap procedure for both selecting the window length and finding critical values. The newly proposed methods are illustrated by applications to astronomical data, to foreign currency exchange rates as well as to water and air pollution data. Finally, the dissertation is concluded by an outlook on further extensions of linearization methods, e.g., in model order selection and change point detection.

This paper is analysing the development of municipal waste amounts in theSwedish municipality of Växjö. The cause of the analysis were decliningwaste amounts in the second half of 2008, which was a rather atypical developmentsince in the years before, a steady growth could be observed.Therefore, the Waste management department in Växjö raised the questionif and in what way the economic downturn that Sweden is currently undergoingcould affect the waste amounts. This is the central question I try toanswer in this paper. To pursue a systematic analysis, I start with a theoreticalsection on how waste is generated in societies, how waste is managed inSweden and which factors influence the development of municipal waste.Aim of this section is to give a theoretical assessment on how the economicdownturn could affect the actual waste amounts. Then, an empirical analysisof Växjö’s municipal waste amounts is conducted. This is done by timeseries modelling of monthly amounts, by using the ARIMA methodology.The models then are tested on structural breaks that could be attributed to aneconomic downturn. Finally, the waste amounts for the next twelve monthsare forecasted.