by Haksun Li | Apr 26, 2015 |

GLASSOFAST is the Graphical LASSO algorithm to solve the covariance selection problem. References “Sustik, M.A. and Calderhead, B., “GLASSOFAST: An efficient GLASSO implementation,” UTCS Technical Report TR-12-29, November 6, 2012.” “O. Banerjee, L. E. Ghaoui and A. d’Aspremont, “Model Selection Through Sparse Maximum Likelihood Estimation for multivariate Gaussian or Binary Data,” Journal of Machine Learning Research, 9, pp. 485-516, March 2008.”...
by Haksun Li | Apr 8, 2015 |

AlgoQuant is a collection of trading-bot building tools. It is an actively developed and ever more comprehensive library of code for building systems for algorithmic trading in all aspects. You can use it to build systems for backtesting, trading strategy generation, data analysis, research, and actual order execution. Out of the box, AlgoQuant is an indispensable and a very powerful tool for quantitative trading research because it is backed by an extensive math library (SuanShu). You are able to very quickly prototype very complicated quantitative strategies by reusing the AlgoQuant components. Among other things that AlgoQuant does are in-sample calibration and optimization, out-of-sample backtesting, performance analysis and strategy generation. Unified Mathematical Analysis and Modeling Library AlgoQuant differentiates itself from other trading system by focusing on quantitative trading strategy research (not execution). (However, there is nothing to prevent you from replacing our mock market with a real market execution/brokerage API.) Built on SuanShu, you can do all the data analysis and sophisticated mathematics based strategies all in Java, without the need to resort to external math software like Excel, R, or Matlab. Additionally, AlgoQuant comes with some well established models published in academic journal papers. Uniqueness Numerical Method Inc. has a unique way of doing backtest. In general, backtest using historical data alone is not a sufficient justification of a quantitative trading strategy. Historical backtest tells only how much money we could have made in the past. What we really want to know is how much money we may make in the future. The future P&L is a random variable. Therefore, in addition to reporting the historical P&L, a proper...
by Haksun Li | Apr 10, 2015 |

Quantitative trading is when traders design mathematical models to describe and predict market movements. Often these models are implemented on computer systems for automatic execution. Traders start with a vague trading intuition. Using mathematics, they turn the idea into a quantitative model for analysis, back testing and refinement. When this quantitative model proves to be likely profitable by passing the rigorous statistical tests, the traders implement them on computer systems for execution. Algorithmic trading is the process where ideas are turned into mathematical models and then coded into computer programs for systematic trading. It is a science where mathematics and computer sciences meet. Here is a collection of resources. Tools AlgoQuant Quantitative Trading Strategies AlgoQuant Quantitative Models mean reversion trend following? NTU 2011 group projects NUS 2011 group projects Research Introduction to Algorithmic Trading Strategies literature historical data...
by henrymorco | Apr 13, 2015 |

After the initial setup, you can run demo programs included in the sub-project algoquant-core. Source files that end in “…Demo.java” (e.g, RunAllDemo.java), are executable programs for demonstration. To run a demo program, you right click on it and hit Run File (or...
by sanjay | Apr 13, 2015 |

Installation On your web site I can read, suanshu.net-2.0.0.zip and algoquant-0.5.0.zip. Should we download both, or does AlgoQuant suffice and we get SuanShu with the algoquant-0.x.x.zip? You just need to download algoquant-0.5.0.zip. It comes with the latest version of SuanShu as well. SuanShu.net is our experimental C# library and is free to download and use. Which version of Java do you recommend? We in house test all our code using a few thousands unit tests daily with 4 versions of jvm from Sun/Oracle: 1.6.x 32 and 64 bits, 1.7.x 32 and 64 bit. Your choice of jvm version is a personal decision. Although some of our developers use the IBM clone with eclipse, it is not part of our routine testing procedure. My personal preference is the Oracle 1.6.x 64 bit. We need 64 bit to process a large amount of data. We prefer 1.6.x because it is a much more mature product with history esp. amid-st the infamous security patches and testing/release policy from Oracle lately. mvn compile does not work! Please do mvn clean install before compile. When building AlgoQuant with Maven, I got errors with the plugin Surefire. E.g., Failed to execute goal org.apache.maven.plugins:maven-surefire-plugin:2.14:test (default-test) on project algoquant-core: Execution default-test of goal org.apache.maven.plugins:maven-surefire-plugin:2.14:test failed This problem seems to happen mostly on a Mac. Please try to use Surefire version 2.14.1 or above. You may try to run the maven build with option “-e”. You may try to skip the unit testing by option “-skipTests”. (In NetBeans, there is such an option under “Maven”.) If none of the above works, our wild guess is that your VM...
## Recent Comments