Martin Magris

TL;DR: In this article, the authors presented the first publicly available benchmark dataset of high-frequency limit order markets for mid-price prediction, where normalized data representations of time series data for five stocks from the NASDAQ Nordic stock market for a time period of ten consecutive days, leading to a dataset of 4,000,000 time series samples in total.

TL;DR: This work investigates the effectiveness of two multilinear methods for the mid-price prediction problem against other existing methods and shows that by utilizing tensor representation, multil inear models outperform vector-based approaches and other competing ones.

TL;DR: This paper describes the first publicly available benchmark dataset of high-frequency limit order markets for mid-price prediction, extracting normalized data representations of time series data for five stocks from the NASDAQ Nordic stock market for a time period of ten consecutive days, leading to a dataset of ~4,000,000 time series samples in total.

TL;DR: An experimental protocol is defined that can be used in order to evaluate the performance of related research methods and baseline results based on linear and nonlinear regression models are provided to show the potential that these methods have for mid-price prediction.

TL;DR: In this paper, the authors describe a new benchmark dataset of high-frequency limit order markets for mid-price prediction and define an experimental protocol that can be used in order to evaluate the performance of related research methods.