32 research outputs found
Time warping algorithms and its applications on financial time series
No full textWe introduce some of the methods for time warping, which is a technique normally used in speech recognition. Discrete time warping genetic algorithm (dTWGA) is a method based on genetic algorithm, which has been commonly used in solving optimization problems when the solution space is large and when there is no analytic form for such solution. Another method, known as dynamic time warping (DTW), makes use of dynamic programming and involves additional constraints compared to dTWGA. We illustrates the use of dTWGA on construction of financial networks. We then apply DTW on financial time series for the purpose of portfolio management. In addition to time warping techniques, we also make use of signal detection theory and concepts borrowed from fuzzy set theory in incorporating technical patterns or chart patterns used by traders and technical analysts into some objective trading strategies in a quantitative approach as contrasted to the usual practice by traders which can be seen as a subjective and qualitative approach in predicting the trend of price.</p
Evolution of financial network through non-linear coupling of time series
No full textThe structure of financial market is captured using an analysis of non-linear coupling between various stocks using a novel time warping method known as discrete time warping genetic algorithm (dTWGA). In contrast to previous studies which estimate the correlations between different time series, dTWGA can be used to analyse time series with different lengths and with data sampled unevenly. Moreover, since the coupling between different time series or at different periods of time would be changing over time, the time delay for the influence of a time series to reach another time series is generally non-linear and time dependent, which would not be well captured with correlation measurements. Our time warping method provides an alternative to overcome this problem and we apply dTWGA on Dow Jones Index and Hang Seng Index and their constituent stocks. Through dTWGA, the coupling between the stock time series provides a network description of the financial market. We perform different measurements of the resultant financial networks to observe the evolution of their topological structure. We observe consistent major topological changes during market crashes, leading to a significant decrease in the size of the network. We expect these technical analyses provide new insights into the systemic risk of financial market in the perspective of the stability of the corresponding network
Optimization of systemic stability of directed network using genetic algorithm
No full textFlow dynamics in directed network can lead to cascade failures from node and link removal, and this is used as a paradigm for systemic risks in financial systems where the flow is a money flow. In order to reduce systemic risk, we analyze the network topology and find ways of rewiring to ensure that the time for the first node failure can be maximized. The analysis is numerical using genetic algorithm to evolve a network by rewiring towards one with higher systemic stability. The results show that a network can become more systemic stable if the incoming flow of all the nodes becomes more similar to that of the outgoing flow. For financial network, the way to reduce the risk of cascade bankruptcies is to share the systemic risk in the form of the fluctuation of capital value transfer by all banks. Our simple model of directed network shows that one way to improve the systemic stability of a network is to rewire it towards a perfect Watts-Strogatz network.</p
ZnO nanorod/GaN light-emitting diodes : the origin of yellow and violet emission bands under reverse and forward bias
No full textAuthor name used in this publication: Xinyi ChenAuthor name used in this publication: Alan Man Ching NgAuthor name used in this publication: Aleksandra B DjurišićAuthor name used in this publication: Kok Wai CheahAuthor name used in this publication: Patrick Wai Keung Fong2011-2012 > Academic research: refereed > Publication in refereed journalVersion of RecordPublishedVoR allowe
Time warping of apneic ECG signals using genetic algorithm
No full textWe construct a method of time warping in quasiperiodic time series analysis using genetic algorithm in order to extract the instantaneous phase difference between a template signal and a testing signal. Contrast to previous studies, which involves correlation estimations to determine the shape similarity of two signals taken from the quasiperiodic time series, time warping perform the comparison of the two signals by first constructing a discrete set of M points formed from uniformly sampled values of the template signal f(t). The discrete set of sample values of the testing signal, g(t'), which contains N points, will be interpolated to form a continuous function so that the difference between the template signal at those M points and the corresponding testing signals are minimize to best preserve the mapping of the two signals. The result of this optimization procedure produces a phase shift function that relates the time t' in the testing signal to the time t in the template signal. Due to the numerous choices in the partitioning of the time domain of the two signals, genetic algorithm is found to be effective in extracting this phase shift function. We apply this theoretical tool of time warping using genetic algorithm to analyze the electrocardiographic (ECG) signals, with the aim to investigate if central apneic and obstructive apneic episodes can be differentiated from non-apneic episodes. Detailed statistical analysis of the phase shift from real ECG data of sleep apnea patient indicates that the difference of both magnitude and phase of the signals can be used to differentiate apneic events from non-apneic events.</p
CUHK electronic theses & dissertations collection
No full textNg, Tsz Ching.Thesis M.Phil. Chinese University of Hong Kong 2014.Includes bibliographical references (leaves 87-92).Abstracts also in Chinese.Title from PDF title page (viewed on 29, September, 2016)
