Predicting a flash crash in the yen/dollar foreign exchange market

ABSTRACT

I examine whether the volume-synchronized probability of informed trading (VPIN) can predict a flash crash in the yen/dollar foreign exchange market. The results show that VPIN using bulk volume classification predicted a recent event. However, VPIN using order flows, which are the amount of the ask-side transaction minus those of bid-side, does not.

KEYWORDS:

• Bulk volume classification
• flash clash
• order flow
• VPIN
• yen/dollar foreign exchange market

JEL CLASSIFICATION:

• F31
• G12
• G14

Acknowledgement

This work is supported by JSPS KAKENHI Grant Number 15K03558 and a grant-in-aid from the Zengin Foundation for Studies on Economics and Finance.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

¹ I also consider 1/20 of median of daily trading volume, and the main result remains unchanged.

Additional information

Funding

This work is supported by JSPS KAKENHI Grant Number 15K03558 and a grant-in-aid from the Zengin Foundation for Studies on Economics and Finance.