An image-to-class dynamic time warping approach for both 3D static and trajectory hand gesture recognition

Schneider

2020

Healthcare

This paper forecasts the future spread of COVID-19 by exploiting the identified lead-lag effects between different countries. Specifically, we first determine the past relation among nations with the aid of dynamic time warping. This procedure allows an elastic adjustment of the time axis to find similar but phase-shifted sequences. Afterwards, the established framework utilizes information about the leading country to predict the Coronavirus spread of the following nation. The presented methodology is applied to confirmed Coronavirus cases from 1 January 2020 to 28 March 2020. Our results show that China leads all other countries in the range of 29 days for South Korea and 44 days for the United States. Finally, we predict a future collapse of the healthcare systems of the United Kingdom and Switzerland in case of our explosion scenario.

Section: Dynamic Time Warpingmentioning

confidence: 99%

Epidemiology of Coronavirus COVID-19: Forecasting the Future Incidence in Different Countries

Schneider

2020

Healthcare

“…Originally, it is used within the framework of spoken word recognition, i.e., the technique eliminates non-linear time shifts between two speech patterns caused by different speaking rates (Juang, 1984;Rath and Manmatha, 2003;Muda et al, 2010). In recent times, dynamic time warping is especially utilized in gesture recognition (Arici et al, 2014;Cheng et al, 2016), chemistry (Jiao et al, 2014;Dupas et al, 2015), and medicine (Rakthanmanon et al, 2012;Fu et al, 2017). Surprisingly, there exist only two academic studies in the context of statistical arbitrage trading.…”

Section: Introductionmentioning

confidence: 99%

Statistical arbitrage with optimal causal paths on high-frequency data of the S&P 500

2018

Quantitative Finance

This paper develops the optimal causal path algorithm and applies it within a fully-fledged statistical arbitrage framework to minute-by-minute data of the S&P 500 constituents from 1998 to 2015. Specifically, the algorithm efficiently determines the optimal non-linear mapping and the corresponding lead-lag structure between two time series. Afterwards, this study explores the use of optimal causal paths as a means for identifying promising stock pairs and for generating buy and sell signals. For this purpose, the established trading strategy exploits information about the leading stock to predict future returns of the following stock. The value-add of the proposed framework is assessed by benchmarking it with variants relying on classic similarity measures and a buy-and-hold investment in the S&P 500 index. In the empirical back-testing study, the trading algorithm generates statistically and economically significant returns of 54.98 percent p.a. and an annualized Sharpe ratio of 3.57 after transaction costs. Returns are well superior to the benchmark approaches and do not load on any common sources of systematic risk. The strategy outperforms in the context of cryptocurrencies even in recent times due to the fact that stock returns contain substantial information about the future bitcoin returns.

“…Initially, it is employed in speech recognition to eliminate non-linear time shifts between two speech patterns as a result of different pronunciation [10][11][12]. Recently, dynamic time warping is used mainly in the field of gesture recognition [13,14], chemistry [15,16], medicine [17][18][19], and finance [9,20,21]. Surprisingly, there exists only one academic study that provides a non-parametric methodology to avoid the criticism of arbitrariness and data sniffing: ref.…”

Section: Introductionmentioning

confidence: 99%

How to Identify Varying Lead–Lag Effects in Time Series Data: Implementation, Validation, and Application of the Generalized Causality Algorithm

Adler

2020

Algorithms

This paper develops the generalized causality algorithm and applies it to a multitude of data from the fields of economics and finance. Specifically, our parameter-free algorithm efficiently determines the optimal non-linear mapping and identifies varying lead–lag effects between two given time series. This procedure allows an elastic adjustment of the time axis to find similar but phase-shifted sequences—structural breaks in their relationship are also captured. A large-scale simulation study validates the outperformance in the vast majority of parameter constellations in terms of efficiency, robustness, and feasibility. Finally, the presented methodology is applied to real data from the areas of macroeconomics, finance, and metal. Highest similarity show the pairs of gross domestic product and consumer price index (macroeconomics), S&P 500 index and Deutscher Aktienindex (finance), as well as gold and silver (metal). In addition, the algorithm takes full use of its flexibility and identifies both various structural breaks and regime patterns over time, which are (partly) well documented in the literature.