A comparison of the forecasting ability of immediate price impact models

Research output: Contribution to journalArticleResearchpeer-review

Abstract

As a consequence of recent technological advances and the proliferation of algorithmic and high-frequency trading, the cost of trading in financial markets has irrevocably changed. One important change, known as price impact, relates to how trading affects prices. Price impact represents the largest cost associated with trading. Forecasting price impact is very important as it can provide estimates of trading profits after costs and also suggest optimal execution strategies. Although several models have recently been developed which may forecast the immediate price impact of individual trades, limited work has been done to compare their relative performance. We provide a comprehensive performance evaluation of these models and test for statistically significant outperformance amongst candidate models using out-of-sample forecasts. We find that normalizing price impact by its average value significantly enhances the performance of traditional non-normalized models as the normalization factor captures some of the dynamics of price impact.

Original languageEnglish
Pages (from-to)898-918
Number of pages21
JournalJournal of Forecasting
Volume36
Issue number8
DOIs
Publication statusPublished - Dec 2017

Keywords

  • Out-of-sample forecasting
  • Price impact
  • Trading costs

Cite this

@article{3e5f193d381549b791340ede5299e79f,
title = "A comparison of the forecasting ability of immediate price impact models",
abstract = "As a consequence of recent technological advances and the proliferation of algorithmic and high-frequency trading, the cost of trading in financial markets has irrevocably changed. One important change, known as price impact, relates to how trading affects prices. Price impact represents the largest cost associated with trading. Forecasting price impact is very important as it can provide estimates of trading profits after costs and also suggest optimal execution strategies. Although several models have recently been developed which may forecast the immediate price impact of individual trades, limited work has been done to compare their relative performance. We provide a comprehensive performance evaluation of these models and test for statistically significant outperformance amongst candidate models using out-of-sample forecasts. We find that normalizing price impact by its average value significantly enhances the performance of traditional non-normalized models as the normalization factor captures some of the dynamics of price impact.",
keywords = "Out-of-sample forecasting, Price impact, Trading costs",
author = "Pham, {Manh Cuong} and Duong, {Huu Nhan} and Paul Lajbcygier",
year = "2017",
month = "12",
doi = "10.1002/for.2405",
language = "English",
volume = "36",
pages = "898--918",
journal = "Journal of Forecasting",
issn = "0277-6693",
publisher = "Wiley-Blackwell",
number = "8",

}

A comparison of the forecasting ability of immediate price impact models. / Pham, Manh Cuong; Duong, Huu Nhan; Lajbcygier, Paul.

In: Journal of Forecasting, Vol. 36, No. 8, 12.2017, p. 898-918.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - A comparison of the forecasting ability of immediate price impact models

AU - Pham, Manh Cuong

AU - Duong, Huu Nhan

AU - Lajbcygier, Paul

PY - 2017/12

Y1 - 2017/12

N2 - As a consequence of recent technological advances and the proliferation of algorithmic and high-frequency trading, the cost of trading in financial markets has irrevocably changed. One important change, known as price impact, relates to how trading affects prices. Price impact represents the largest cost associated with trading. Forecasting price impact is very important as it can provide estimates of trading profits after costs and also suggest optimal execution strategies. Although several models have recently been developed which may forecast the immediate price impact of individual trades, limited work has been done to compare their relative performance. We provide a comprehensive performance evaluation of these models and test for statistically significant outperformance amongst candidate models using out-of-sample forecasts. We find that normalizing price impact by its average value significantly enhances the performance of traditional non-normalized models as the normalization factor captures some of the dynamics of price impact.

AB - As a consequence of recent technological advances and the proliferation of algorithmic and high-frequency trading, the cost of trading in financial markets has irrevocably changed. One important change, known as price impact, relates to how trading affects prices. Price impact represents the largest cost associated with trading. Forecasting price impact is very important as it can provide estimates of trading profits after costs and also suggest optimal execution strategies. Although several models have recently been developed which may forecast the immediate price impact of individual trades, limited work has been done to compare their relative performance. We provide a comprehensive performance evaluation of these models and test for statistically significant outperformance amongst candidate models using out-of-sample forecasts. We find that normalizing price impact by its average value significantly enhances the performance of traditional non-normalized models as the normalization factor captures some of the dynamics of price impact.

KW - Out-of-sample forecasting

KW - Price impact

KW - Trading costs

UR - http://www.scopus.com/inward/record.url?scp=84995621430&partnerID=8YFLogxK

U2 - 10.1002/for.2405

DO - 10.1002/for.2405

M3 - Article

VL - 36

SP - 898

EP - 918

JO - Journal of Forecasting

JF - Journal of Forecasting

SN - 0277-6693

IS - 8

ER -