17

Stanford University has a free online course in machine learning with video lectures, problem sets, and even a promise of online help with coursework from Stanford faculty. This course provides a broad introduction to machine learning, datamining, and statistical pattern recognition. Topics include: (i) Supervised learning (parametric/non-parametric ...


17

Yes. First, it is much easier to proceed if you standardize the output of your forecast so they are in the same units (returns, for example, or probabilities of an event/condition occurring). After you have done this, there are 3 general approaches: Signal weighting: Then you need to define a weighting scheme for your factors. Richard Grinold has an one ...


16

This is a very interesting question. I believe it is getting a lot of up-votes from people who have wondered the same thing and don't know where to begin, whereas you have at least laid out a reasonable-sounding plan. I commend you for that. However, it is not clear to me what you're trying to learn by posting this question. In my opinion, the plan you ...


13

As the others have already mentioned, this is a very broad question. Anyway, as a starting point there are some blogs that come to my mind that have some up to date high quality content on these issues from time to time: http://quantivity.wordpress.com/ http://epchan.blogspot.com/ http://www.automated-trading-system.com/ http://intelligenttradingtech....


12

Windham Capital Management is using hidden markov models for their Risk Regime Strategies. Mark Kritzman, who is also CEO, has published an article about the general outline of the strategy (with source code so you can replicate the results!): Regime Shifts: Implications for Dynamic Strategies (corrected August 2012) by M. Kritzman, S. Page, D. Turkington]...


11

I think R's CRAN Task Views on Machine Learning is an excellent resource for beginners moving to advanced algorithm traders. It is well-structured, broad, up-to-date, and ready-to-use! http://cran.r-project.org/web/views/MachineLearning.html I believe all advanced quantitative traders already know this. But I haven't seen anyone post it here and Flake's ...


11

This answer summarizes some of my comments. HFT is certainly a very hot topic these days, but it's hard to point to any one reason. A large part of it is the mystery and the profits, but also part of it is the relative novelty. Note that there is no lack of papers about medium and low frequency strategies, it's just that they are not labeled as such. Medium ...


10

I found this solid overview of different trading algorithms by Deutsche Bank Research: Trade execution algorithms Designed to minimise the price impact of executing trades of large volumes by ‘shredding’ orders into smaller parcels and slowly releasing these into the market. Strategy implementation algorithms Designed to read real-time market data and ...


9

We cannot give you a relative bid-ask spread that would make sense. The reason for that is that it really depends on several parameters: The type of financial asset you invest in (futures, funds, index, options, ...) The period during which you're trading (I think the liquidity in markets hasn't been the same over time). If you trade intraday, it depends on ...


8

There is a paper of mine answering To this question: Dealing with the Inventory Risk. A solution to the market making problem by Olivier Guéant, Charles-Albert Lehalle, Joaquin Fernandez Tapia.


8

First, we are few quants and academics to use the full toolkit of machine learning: stochastic algorithms, to optimal trading. Here are at least two papers: Optimal split of orders across liquidity pools: a stochastic algorithm approach, Sophie Laruelle (PMA), Charles-Albert Lehalle, Gilles Pagès (PMA) Optimal posting distance of limit orders: a stochastic ...


8

Ha, interesting, so many responses with "negative" expectations. There are plenty of people that have successfully gone down this road and are producing pretty nice returns, so obviously it is possible. A trader with a smaller capital has better chances of producing good ROC with very reasonable risk parameters, simply because he's would not be constrained ...


7

From this site's perspective, I think nothing would be better than a ML.SE. Finally, we got one awhile ago. UPDATE: Unfortunately, Machine Learning is merging into Cross Validated. To learn more detail, click here." I have no idea why SE admin was rush to merge ML into CrossValidated. Not a fan of it (Orz). I personally prefer a separate site. FYI, http:...


7

Measuring expected shortfall (also known as conditional value-at-risk) answers the simpler question of "what is my average expected loss at the i-th quantile?" given the empirical distribution of returns. A variation is value-at-risk which measures the loss at the i-th quantile. Arguably you could leave at this this and you have your answer. You probably ...


7

Personally, I am very skeptical of the claims in "Twitter mood predicts the stock market". There are several other papers with similar claims, but not so much good quality research is available. Arguably, the sweet bits of these approaches are not public. A sounder approach is to dig at the relationship between social media activity and relate it to the ...


7

As a starting point to my answer, I would say that reading a book is not sufficient to start doing automated trading on your own as chrisaycock suggests in his comment. I would answer your questions in 3 different ways. First, building your "AI bot" which I would rather call a systematic algorithm not only requires programming skills, it also means having ...


7

The most commonly-known approach to this is described in Inferring trade direction from intraday data (1991) by Lee and Ready. You will find that the non-trivial part has to do with classifying trades that are reported inside the spread. I believe you will find that the Lee-Ready algorithm will outperform the naive midpoint reference approach suggested by @...


7

If I was in your position I would start to research how I can create a web server is C++ and expose calls to create a REST service. In other words, can you make your code status output to HTTP? From there, the rest should be easy. You would just need to create a GUI that can access REST services, which virtually all modern languages can. You could focus on ...


6

The most basic strategy is beta-based quantiles. That is to say, you first control for losses on your individual stock versus overall market performance. (Your trading strategy may or may not wish to hedge away the market factor using, say, SPX futures). Then you choose a quantile, call it the 5th percentile, beyond which you consider a move to be ...


6

In 2010 Informs and Kaggle organized time series prediction contest. The methods used by competitors are described here.


6

Jordan (@jordan.baucke) in his answer suggests that most latency arbitrages are actually market making strategies, as opposed to classical price arbitrage. While I generally agree, I can think of two exceptions: Equity price arbitrage in fragmented markets (See the fragulator for more on this). In this environment, negative spreads can arise and the ...


6

I assume that by "how much research" you mean "could you provide me with some links".... So, as @Shane mentioned in its comment, a hedge fund recently started and is focusing on twitter analysis (here is another link). From what I understand they are basically implementing a trend-following strategy based on the "mood" of twitter users (I, of course, don't ...


6

I'd say it depends on how close you want to be to reality and what the strategy entails. For instance one scenario when actual currency makes sense is when you want to take contract sizes and position limits into account, for instance agricultural futures contracts nearly always impose a position limit for one party in one or all contracts. If your ...


6

http://www.portfolioprobe.com/2010/11/05/backtesting-almost-wordless/ shows an example of how the results from a backtest can be deceiving. This would be true with either returns or value. The main issue is that the portfolio you start with can have an impact on what "good" means.


6

A quick Google search gives a few hints: http://etf.about.com/od/etfinvestingstrategies/a/ETF_Arbitrage.htm http://ftalphaville.ft.com/blog/2011/05/18/572086/how-profitable-is-etf-arbitrage/ http://www.iijournals.com/doi/abs/10.3905/jii.2010.1.1.107 http://seekingalpha.com/article/68064-arbitrage-opportunities-with-oil-etfs Another quick search on scholar....


6

Concur with Thomas for most part, though I would recommend you to sign up for a trial with Dow Jones Newswire. I like the API and app that Newsware ( http://www.newsware.com/) makes available. It is not suitable for hft but I use it in order to stay informed and look up often used mnemonics. I think they have a pretty capable API and I remember they offer ...


6

The first formal model to explain this was Kyle (1985). Oversimplifying, imagine there is a group uninformed traders and an informed trader active in the market for a given security. Uninformed traders cannot make correct directional predictions. The informed trader knows --- with some uncertainty --- what the price will be at the end of a certain time ...


6

To be honest you're not likely to get a very satisfying answer to your question. Not because its a bad question, but because "regular people" can't just go hooking their home grown trading systems up to a live market. I'd like to start automating my trading strategies. First off you'll need a system that can interface with your broker. If you're not a ...


5

HFT, when they implement market-making like strategies, are a key element of a fragmented market to build "arbitrage bridges" between trading venues. There is a cost that for: we are all paying (probably around a fraction of the actual spread) to them, and the resiliency of the order books suffers because of their presence. As usual, there are positive and ...


5

I can share my own experience working with the Deltix product suite. As a research and development platform it's very feature rich with support for every back-testing mode there is (BBO, Trade, Midprice, Bar, Level 2 Order Book) and advanced optimization modes (walk-forward, genetic, mean-variance, portfolio optimization, etc). I have built components and ...


Only top voted, non community-wiki answers of a minimum length are eligible