Ramsey Theory says that complete disoder is impossible. Now a major result in RT has implications for market efficiency and random walks [N.B. disorder and random walks and efficiency are tricky concepts and seeing the connections between them is not simple] http://www.elitetrader.com/et/index.php?posts/4288640
Probably interesting to several people that read this thread: https://www.quantopian.com/posts/pair-trade-with-cointegration-and-mean-reversion-tests
Microstructure Links ATS Transparency Data – FINRA information on Alternative Trading Systems BATS Market Summary – market share statistics Decimus Capital Markets KOR Group – broker execution data MIDAS – microstructure analytics platform used by the SEC Modern Markets Initiative – HFT industry group Nanex – market data vendor SEC Equity Market Structure Advisory Committee Tabb Forum – forum for industry opinion articles
Very interesting. Thanks for the share. I saw the complex order types. I really don't think we should have more than just LMT and MKT. Why have order types that say "passively follow the bid and cross against market orders, but cancel if there is a sweep order blah blah blah blah"?
Code: CPI Apr May Jun Jul Aug Sept Oct Nov Dec Jan Feb Mar Apr May 0.1 0.3 0.2 0.1 0 -0.1 0.2 0.1 -0.1 0 -0.2 0.1 0.4 ? Import Prices May June July Aug Sept Oct Nov Dec Jan Feb Mar Apr May 1.1 0.1 -0.9 -1.8 -1.1 -0.3 -0.6 -1.2 -1.3 -0.5 0.4 0.7 1.4
Without me turning into a conspiracy theorist and just being a computer scientist, in general terms you want to push the complexity of a system as far down the decision making point as possible. It is far more efficient for the exchange to have to keep track of these complexities, once, instead of every algo developer each time. The exchanges integrity is more likely to suffer less that way. Exchange programmers tend to be very good. Algo system builders less so. Further, generally you want orders that are managed at the server - this reduces latency since you don't have to continuously cancel/replace orders increasing bandwidth needs, and of course it is more fault tolerant. So instead of canceling orders, you are canceling algorithms. Far more efficient, and the right level of abstraction. There are lots of good reasons for many of these order types. I think very very few of these things were meant to be evil. A couple of them definitely are though and definitely don't make for a fair marketplace (which is the #1 poison to an exchange). The exchanges in their quest to compete, let those few slide. Wrong decision.
I see what you mean -- I am all for the PEG or REL order, as it allows a retail to flow the bid without having a computer with HF quotes. In general, I don't mind order types that I can see at Interactive Brokers, which is already a lot more than Fidelity or Etrade. I am talking about the order types that are not even available on IB, and when I look up the NYSE documentation it is NOT even there. They are hidden in the legal documents that no one reads. An exchange needs to have clear rules that everyone understands, because we want to maximize the number of investors, much like credit cards have plain English explanations.
By no-one of course you mean nearly no one. The HFT programmers pour over that stuff. It is incredibly boring (I bore easily), but if you can plow through it, and understand how it can be used to advantage, these people have shown how it can lead to incredible profits at the expense of the buy side.
Copied from an email I got. INTRODUCING MERGENT GLOBAL FUNDAMENTALS Our selection of global fundamentals data continues to expand with the addition of Mergent Global Fundamentals Data. Compiled by established data provider, Mergent, this data provides valuable insight into global markets. It covers more than 10,000 companies from 67 countries as listed in the Russell Global Index, which represents a large majority of the investable universe. HIGHLIGHTS OF THIS DATABASE: Over 50 fundamental indicators directly sourced or derived from company income statements, balance sheets and cash flow statements Data history going back to 2005 Daily updates made by Mergent’s highly experienced team of analysts Reference-grade quality guaranteed thanks to Mergent’s rigorous multi-stage validation process, applied to every datapoint, for unparalleled accuracy