Cryptocurrency Arbitrage

Definition

Cryptocurrency arbitrage can be described as a set of trading operations that is performed simultaneously by buying and selling the same cryptocurrency amount from two different exchanges to gain a profit off the exchange rate difference. The transaction is carried out in a way that the cryptocurrency amount on both exchanges is equal to the amount before the transaction, while the equivalent amount in fiat is increased by the rate difference.

Description

Histrically, crytocurrency arbitrage derives its roots from statistical arbitrage. Statistical arbitrage performed on traditional markets was set to gain profit by simultaneously buying and selling shares of two correlated instruments. It is recomended to find two securities that are in the same sector / industry, they should have similar market capitalization and average volume traded. SA Strategy calculates two following values, which are later used to generate proper trade signals:

  • Mean (M) ratio of the first n periods prices of two assets
  • Standard deviation (SD) of ratio of the first n periods prices of two assets

where n is a Number of Periods.

Mean and standard deviation are defined as follows:

\text{M} = \frac{\sum_{i = 1}^{n}{\frac{A_i}{B_i}}}{n}

\text{SD} = \sqrt{\frac{1}{n-1} \sum_{i = 1}^{n} (\frac{A_i}{B_i} - M)^2}

where A_i (B_i) is a price of the first (second) asset at the moment i.

SA strategy starts decision rule, if orders should be sent after receiving Number of Periods trades.

Market Data

  • Latest Tick

Parameters

PARAMETER NAME DESCRIPTION ESSENTIAL
Number of Periods Number of last ‘last trades’ events taken into account in a computation of the mean and standard deviation Yes

Conditions

Open position

Side Buy or Sell
Amount One asset per order
Price Last market price
Type Market

Consider an arbitrary moment k \geq n, when strategy has no opened positions. If \frac{A_k}{B_k} > M + 2 \times SD” /> then we sell asset A and buy B. Similarly when <img class= then we buy asset A and sell B.

Close position

Consider an arbitrary moment l > k \geq n” />, when strategy has already some opened positions. If we buy one asset and sell the other at <img class= we should close these two positions when \frac{A_l}{B_l} = M.

Termination

There is no strategy termination condition.

Time frame

This strategy is dedicated to run in a long period of time. The instruments are highly correlated, so it is not likely that the ratio of prices will differ from mean by two standard deviations.

Further information & source

  • Barry Johnson, Algorithmic Trading & DMA: An introduction to direct access trading strategies, 4Myeloma Press, 2010.
A complete guide to Algorithmic Trading

Recommended readings:

Basket Orders, Bollinger Bands, CSI, MACD, Market Close, Parabolic SAR, POV, RSI, Slow Stachastic Osilliator, TWAP, VWAP, William %R, Adoptive Shortfall, Triangular Arbitrage