VPIN

Volume Synchronized Probability of Informed Trading

Introduction

• The VPIN method intends to measure the probability of market informed transaction.
• Although VPIN method is viewed for the HFT environment. But the result suggests that certain VPIN model with Linear Regression provides proxies for adverse selection risk
• 주식시장에서의 매수, 매도는 50:50으로 이루어진다는 가정하에 거래량의 불균형을 계산하는 방법입니다.
• 보고자 하는 주식 종목의 일일 평균 거래량을 잘 파악해서 버켓의 크기와 VPIN 계산하기 위한 버캣개수 개수를 잘 설정 해줘야합니다.

Info

• Version - Written in Python 3.7.3
• Keywords
  • VPIN
  • Market micro structure

Prerequisite

• pandas
• numpy
• matplotlib.pyplot
• time
• train_test_split
• LinearRegression

Correspondence

Please do not hesitate to submit an issue or contact via ichuzupikachu@gmail.com.

Usage

dataset

• kakaostock.csv
• Sampling time: 2021-March-17
• Data sample in time period between 1000 and 1400.

Methodology

• For transactions in the sampling date, the entile volume is divided into 1000 buckets.
• Filling Procedure: Filling of bucket starts when transaction starts. When volume of transaction exceeds the upper bound, calculate vol_bucket with buy volume, which indicates the rest of transaction amount Loop of aforementioned process generate a series of baskets.
• 

Sample Output

• Price change per VPIN gradient