Search

Your search keyword '"Tsang, Edward P.K."' showing total 15 results
Start Over Author "Tsang, Edward P.K."
15 results on '"Tsang, Edward P.K."'

5. Directional change for handling tick-to-tick data.

Author

Tsang, Edward P.K.

Abstract

Time Series (TS) records transactions in a market at fixed intervals. Directional Change (DC) is an alternative way to record transactions: it only records transactions that represent significant price changes in the opposite direction in a trend, where 'significance' is observer-defined. In this paper, we argue that DC is particularly suitable for recording and analysing tick-to-tick data. Firstly, significant data points and high activities between sampling points that may not be recorded in a TS will always be recorded in DC. Secondly, as transactions take place at irregular times, but TS records transactions at fixed intervals, adjustments are required in the recording process, which may distort the records; no adjustments are required in DC. Thirdly, as DC is data-driven: every new transaction could potentially provide us with new information on the pulse of the market. For these reasons, DC is more suitable than TS for tracking tick-to-tick data for signals. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

15. DE/EDA: A new evolutionary algorithm for global optimization.

Author

Jianyong Sun, Qingfu Zhang, and Tsang, Edward P.K.

Subjects
*ALGORITHMS, *MATHEMATICAL optimization, *ESTIMATION theory, *ALGEBRA, *DISTRIBUTION (Probability theory)
Abstract

Differential evolution (DE) was very successful in solving the global continuous optimization problem. It mainly uses the distance and direction information from the current population to guide its further search. Estimation of distribution algorithm (EDA) samples new solutions from a probability model which characterizes the distribution of promising solutions. This paper proposes a combination of DE and EDA (DE/EDA) for the global continuous optimization problem. DE/EDA combines global information extracted by EDA with differential information obtained by DE to create promising solutions. DE/EDA has been compared with the best version of the DE algorithm and an EDA on several commonly utilized test problems. Experimental results demonstrate that DE/EDA outperforms the DE algorithm and the EDA. The effect of the parameters of DE/EDA to its performance is investigated experimentally. [ABSTRACT FROM AUTHOR]

Published
2005
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results