TUTCRIS - Tampereen teknillinen yliopisto

TUTCRIS

Benchmark dataset for mid‐price forecasting of limit order book data with machine learning methods

Tutkimustuotosvertaisarvioitu

Standard

Benchmark dataset for mid‐price forecasting of limit order book data with machine learning methods. / Ntakaris, Adamantios; Magris, Martin; Kanniainen, Juho; Gabbouj, Moncef; Iosifidis, Alexandros.

julkaisussa: JOURNAL OF FORECASTING, Vuosikerta 37, Nro 8, 12.2018, s. 852-866.

Tutkimustuotosvertaisarvioitu

Harvard

Ntakaris, A, Magris, M, Kanniainen, J, Gabbouj, M & Iosifidis, A 2018, 'Benchmark dataset for mid‐price forecasting of limit order book data with machine learning methods', JOURNAL OF FORECASTING, Vuosikerta. 37, Nro 8, Sivut 852-866. https://doi.org/10.1002/for.2543

APA

Vancouver

Author

Ntakaris, Adamantios ; Magris, Martin ; Kanniainen, Juho ; Gabbouj, Moncef ; Iosifidis, Alexandros. / Benchmark dataset for mid‐price forecasting of limit order book data with machine learning methods. Julkaisussa: JOURNAL OF FORECASTING. 2018 ; Vuosikerta 37, Nro 8. Sivut 852-866.

Bibtex - Lataa

@article{e222c7f2b52b4734a5417fe4d9c2744d,
title = "Benchmark dataset for mid‐price forecasting of limit order book data with machine learning methods",
abstract = "Managing the prediction of metrics in high‐frequency financial markets is a challenging task. An efficient way is by monitoring the dynamics of a limit order book to identify the information edge. This paper describes the first publicly available benchmark dataset of high‐frequency limit order markets for mid‐price prediction. We extracted normalized data representations of time series data for five stocks from the Nasdaq Nordic stock market for a time period of 10 consecutive days, leading to a dataset of ∼4,000,000 time series samples in total. A day‐based anchored cross‐validation experimental protocol is also provided that can be used as a benchmark for comparing the performance of state‐of‐the‐art methodologies. Performance of baseline approaches are also provided to facilitate experimental comparisons. We expect that such a large‐scale dataset can serve as a testbed for devising novel solutions of expert systems for high‐frequency limit order book data analysis.",
author = "Adamantios Ntakaris and Martin Magris and Juho Kanniainen and Moncef Gabbouj and Alexandros Iosifidis",
note = "EXT={"}Iosifidis, Alexandros{"}",
year = "2018",
month = "12",
doi = "10.1002/for.2543",
language = "English",
volume = "37",
pages = "852--866",
journal = "JOURNAL OF FORECASTING",
issn = "0277-6693",
publisher = "Wiley",
number = "8",

}

RIS (suitable for import to EndNote) - Lataa

TY - JOUR

T1 - Benchmark dataset for mid‐price forecasting of limit order book data with machine learning methods

AU - Ntakaris, Adamantios

AU - Magris, Martin

AU - Kanniainen, Juho

AU - Gabbouj, Moncef

AU - Iosifidis, Alexandros

N1 - EXT="Iosifidis, Alexandros"

PY - 2018/12

Y1 - 2018/12

N2 - Managing the prediction of metrics in high‐frequency financial markets is a challenging task. An efficient way is by monitoring the dynamics of a limit order book to identify the information edge. This paper describes the first publicly available benchmark dataset of high‐frequency limit order markets for mid‐price prediction. We extracted normalized data representations of time series data for five stocks from the Nasdaq Nordic stock market for a time period of 10 consecutive days, leading to a dataset of ∼4,000,000 time series samples in total. A day‐based anchored cross‐validation experimental protocol is also provided that can be used as a benchmark for comparing the performance of state‐of‐the‐art methodologies. Performance of baseline approaches are also provided to facilitate experimental comparisons. We expect that such a large‐scale dataset can serve as a testbed for devising novel solutions of expert systems for high‐frequency limit order book data analysis.

AB - Managing the prediction of metrics in high‐frequency financial markets is a challenging task. An efficient way is by monitoring the dynamics of a limit order book to identify the information edge. This paper describes the first publicly available benchmark dataset of high‐frequency limit order markets for mid‐price prediction. We extracted normalized data representations of time series data for five stocks from the Nasdaq Nordic stock market for a time period of 10 consecutive days, leading to a dataset of ∼4,000,000 time series samples in total. A day‐based anchored cross‐validation experimental protocol is also provided that can be used as a benchmark for comparing the performance of state‐of‐the‐art methodologies. Performance of baseline approaches are also provided to facilitate experimental comparisons. We expect that such a large‐scale dataset can serve as a testbed for devising novel solutions of expert systems for high‐frequency limit order book data analysis.

U2 - 10.1002/for.2543

DO - 10.1002/for.2543

M3 - Article

VL - 37

SP - 852

EP - 866

JO - JOURNAL OF FORECASTING

JF - JOURNAL OF FORECASTING

SN - 0277-6693

IS - 8

ER -