1.
해외 알고리즘트레이딩 소스를 보면 IB API를 많이 사용합니다.IB API가 무슨 특별한 물건인 것이 아닐까 생각할 수 있지만 국내 증권사들이 제공하는 HTS API나 DMA API와 다르지 않습니다. 기능이나 구조가 다르지만 시세와 매매를 제공하는 본질은 같습니다.
여기 두개의 논문이 있습니다. 첫번째 논문은 ‘Developing high-frequency equities trading model’입니다.
The purpose of this paper is to show evidence that there are opportunities to generate alpha in the high frequency environment of the US equity market, using Principal Component Analysis (PCA hereafter) as a basis for short term valuation and market movements prediction. The time frame of trades and holding periods we are analyzing oscillate between one second to as high as 5 minutes approximately.
두번째 논문은 ‘Profiting from mean-reverting yield-curve trading strategies’입니다.
위의 두 논문을 기초로 하여 알고리즘을 만들면 어떤 모양일까요?
2.
두 논문을 기초로 하여 아래와 같은 특징을 가진 통계적 차익거래모델을 만들어 보자고 하죠.개발언어는 Python입니다.
- Cash-neutral strategy with long-short position
- Bootstrap the model with historical data to derive usable strategy parameters
- Bootstrapping takes some time, we need to bridge historical data with recent tick data
- Transforming inhomogenous to homogeneous time series of 1 second intervals
- Selection of highly-correlated stock pairs
- Using volatility ratio to detect trending, mean-reversion or Brownian motion
- Fair valuation by using beta of average 5 minute look-back price window
- Fair valuation of stock A against more than 1 security (stock B, C…) is possible
- Trade decisions based on mean-reversion, volatility ratio and deviation from fair prices
- Generate trade signals and place buy/sell orders based on every incoming tick data
- Re-evaluating beta every 30 seconds to account for small regime shifts
할 수 있을까요? 이런 흐름을 교육을 하고 알고리즘을 직접 개발하는 교육과정이 필요하다는 생각을 했고 이런 고민을 담았던 글이 아래입니다.
아직까지 현실화하지 못하고 있습니다. 저의 능력을 벗어난 부분이 있습니다. 그래도 계속 꿈을 꿉니다. 다시 처음으로 돌아가서 결과물을 확인하죠. Python으로 개발한 알고리즘이 아래입니다. 이전에 소개했습니다.
3.
얼마전 Tabb Forum이 소개한 논문입니다.
This paper builds a model of high-frequency equity returns by separately modeling the dynamics of trade-time returns and trade arrivals. Our main contributions are threefold. First, we characterize the distributional behavior of high-frequency asset returns both in ordinary clock time and in trade time. We show that when controlling for pre-scheduled market news events, trade-time returns of the highly liquid near-month E-mini S&P 500 futures contract are well characterized by a Gaussian distribution at very fine time scales. Second, we develop a structured and parsimonious model of clock-time returns by subordinating a trade-time Gaussian distribution with a trade arrival process that is associated with a modified Markov-Switching Multifractal Duration (MSMD) model. This model provides an excellent characterization of high-frequency inter-trade durations. Over-dispersion in this distribution of inter-trade durations leads to leptokurtosis and volatility clustering in clock-time returns, even when trade-time returns are Gaussian. Finally, we use our model to extrapolate the empirical relationship between trade rate and volatility in an effort to understand conditions of market failure. Our model suggests that the 1,200 km physical separation of financial markets in Chicago and New York/New Jersey provides a natural ceiling on systemic volatility and may contribute to market stability during periods of extremely heavy trading.
A Compound Multifractal Model for High-Frequency Asset Returns
Information Transmission between Financial Markets in Chicago and New York
