Optimal Pairs Trading Strategy under Geometric Brownian Motion and its Application to the US stocks

This study is a study on pair trading, a representative market-neutral investment strategy. A general pair trading strategy uses econometric techniques to select a pair of stocks and calculates the trading price level depending on a single variable called the variance of stock returns without any theoretical background. This study applies the optimal pair trading strategy proposed by Liu et al. (2020) to the top US market cap stocks and examines its performance. This strategy proposes a mathematical background for optimally calculating the trading price level. Since the statistical method for pair selection can be omitted, a pair can be formed only with good stocks with guaranteed liquidity. In addition, strategic risk management is possible because the stop loss set according to the market situation is performed. As the top 10 market cap stocks traded on the US exchange, daily closing price data for 10 years from 2011 to 2020 were applied to optimal pair trading. It was confirmed that the rate of return may differ depending on the adjustment of various parameters including the level of stop loss. In this study, an applicated strategy that properly managed pairs trading and stocks together earned the minimum annual average return 17.88% and the Sharpe ratio reached 1.81. These numbers can be better with the adjustment of the parameters.

OPTIMAL PAIR-WISE SUS ALGORITHM AS PROPORTIONAL FAIR ALGORITHM FOR MU-MIMO-OFDM SYSTEMS

In this paper, we research proportional fairness of the optimal pair-wise semi-orthogonal user selection (SUS) algorithm used in the MU-MIMO-OFDM system compliant with IEEE 802.11ac standard. Zero-forcing beamforming (ZFBF) is applied in the system as a precoding technique in order to cancel user interferences. Two different scenarios are considered: homogeneous and heterogeneous. Beside the proportional fairness, a throughput performance of the MU-MIMO-OFDM system is analyzed for both scenarios too. Simulation results show that the optimal pair-wise SUS algorithm is an excellent solution to realize the trade-off between the proportional fairness among users and the achieved system throughput.

Serial and parallel implementation of Needleman-Wunsch algorithm

Needleman-Wunsch dynamic programming algorithm measures the similarity of the pairwise sequence and finds the optimal pair given the number of sequences. The task becomes nontrivial as the number of sequences to compare or the length of sequences increases. This research aims to parallelize the computation involved in the algorithm to speed up the performance using CUDA. However, there is a data dependency issue due to the property of a dynamic programming algorithm. As a solution, this research introduces the heterogeneous anti-diagonal approach, which benefits from the interaction between the serial implementation on CPU and the parallel implementation on GPU. We then measure and compare the computation time between the proposed approach and a straightforward serial approach that uses CPU only. Measurements of computation times are performed under the same experimental setup and using various pairwise sequences at different lengths. The experiment showed that the proposed approach outperforms the serial method in terms of computation time by approximately three times. Moreover, the computation time of the proposed heterogeneous anti-diagonal approach increases gradually despite the big increments in sequence length, whereas the computation time of the serial approach grows rapidly.