Tag Info

Hot answers tagged

21

As mentioned elsewhere on this site, Lo, Mamaysky, and Wang (2000) do exactly what you're talking about, namely algorithmic detection of head and shoulders patterns. Their definition: Head-and-shoulders (HS) and inverted head-and-shoulders (IHS) patterns are characterized by a sequence of five consecutive local extrema $E_1,...,E_5$ such that $$ HS ...


12

I would recommend that you read "Evidence-Based Technical Analysis" by David Aronson. Firstly, I am mentioning it because it is a highly worthwhile book. Secondly, on pp151--161 he attempts to "objectify subjective TA", using the head-and-shoulders pattern as an example.


12

Seeing a pattern in a chart is the finance equivalence of a Rorschach test---the discerned pattern says more about the person than the image. And really, if you want to trade that way, you may as well use astrology. Your real question seems to be: How can I accept or reject the hypothesis that Bollinger bands are an acceptable trading signal? For ...


12

Below, I see a lot of support and resistance. Here's the code: x <- cumsum(rnorm(1000)) plot(x, type="l", main="Support and Resistance") Edit (03/03/2011) ================================================ Gortaur, I put my answer here to avoid filling up the comment area. Your question 1) "......I was not asking for the "garbage" literature, I can ...


10

Most contemporary NN systems are just made to use the raw price time series for input (maybe with some kind of simple normalization), but for my thesis I wrote a system which traded equities with an ANN with technical indicator inputs (MAs, MACD, even pattern matching for stuff like Head-Shoulders, support levels, etc.). So at least conceptually it's ...


9

There is certainly much more to quantitative finance than technical analysis, and a previous question does a decent job of outlining the different areas, as does the wikipedia on "quantitative analyst". Even for what wikipedia terms an "algorithmic trading quant" or what Mark Joshi terms a "statistical arbitrage quant", technical analysis is just one tool ...


8

The predictor variables would consist of the input layer to the neural network. The output layer would consist of your target. You need to specify the hidden layer, number of nodes per layer, the learning algorithm, and the learning algorithm stopping criteria. Typically inputs are normalized (first-differenced, z-scored, etc.) before inputting into the ...


7

The only "indicators" that I believe add value in academic research are time series smoothing functions. ( I don't call them indicators because they are all lagging thus do not indicate anything into the future). There is clear empirical evidence and a number of academic papers have been published that show that none of the common indicators (common ...


6

I think the answer to your question is very dependent on the respective indicators. It can be argued for example that moving averages not only smooth out time series but because they are a shifted version of the original series signals on crossovers make use of the momentum factor. In general you might want to check out the book Evidence Based Technical ...


5

Hi Quantitative Finance has in my opinion two main streams. The first is about of valuation of some derivative contracts in a consistent way. This is a theory and once paradigms accepted it is coherent, it can considered as science at the same level as economy can pretend to this kind of terminology. The second is about making (or trying to) prediction(s) ...


5

This entire approach hinges on how you define "value." Once you've defined value, you can define a metric for "stability" or "risk." A working hypothesis would be that stock that have been stably valuable in the past would continue to be valuable in the future. Of course, this is a hypothesis. Let's say (for sake of an example, this is not financial ...


5

Dynamic Time Warping, recursive, time-delayed feedforward neural networks, wavelets, empirical mode decomposition, ..., there's plenty of it. BUT If you want my advice, don't go this way, I wasted too much time doing things like that. Neither big nor small players (profitably and consistently) trade this way and for a good reason. Technical analysis is a ...


5

Contrary to popular belief, there does exist some truly high quality academic literature on this topic. The most sophisticated and well executed paper in this regard is Lo, Mamaysky, and Wang (2000). They write: In this paper, we propose a systematic and automatic approach to technical pattern recognition using nonparametric kernel regression, and we ...


5

Just like everyone else that's been down this path, you'll have to prove this stuff to yourself. Make sure that one of your competing tests is a "noise test" where the decision to go long or short is driven by a meaningless random number generator. If your method can't statistically outperform noise, then your method is not doing anything meaningful.


5

Remember that there is almost no point in predicting market movements if you cannot use it to trade and generate P&L. Thus, backtesting a stat arb strategy based on the indicator is best option. Don't let yourself fooled by correlation or even directional forecast percentage accuracy as a few wrong predictions can blow your capital. You will need a ...


5

I will break up your question in to some parts to make answering easier. "people use various economic indicators with their networks (moving average, MACD, etc.) However, how do these come into play in a NN context?"--the 'indicators' MA, MACD etc. come from the data. They are measures of the data capturing some aspect. You could try to capture/replicate ...


4

There is so much finance literature on this topic, I don't even know where to begin. Specifically on momentum, some of the earlier foundational papers are Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency Momentum Strategies Price Momentum and Trading Volume International Momentum Strategies Momentum has an entire ...


4

You might want to check out the book Evidence Based Technical Analysis by David Aronson. In it he applies statistical techniques to determine whether certain technical analysis indicators and ensembles have any predictive power. It's an interesting read and should equip you with some ideas on how you might perform a similar analysis.


4

Quant in trading creates system that can be backtested, has a certain risk valuation. It is more like playing chess when you need to calculate multistep strategy. Let say certain instrument moves 1% a day. Our goal is to create strategy for one year (250 step strategy). If we use stock + options we get 50 or more entries a day into our system for analysis. ...


3

It seems that this is the key difference between OBV and TSV: "Time segmented volume is the way to get consistent volume data and eliminate all the volume distortions that we discussed above. Here's the key to why time segmented volume works: Let's start with volume on a 5 minute chart and for this example, look at the 10:15 bar. Now take the average of ...


3

I implemented these algorithms just a few months ago! I would just double the number of data points that you need. If you need a 10 day EMA, take 20 data points. For the MACD, you need the EMA, so add them together then double it. Here is the algorithm including an Excel spreadsheet you can compare with: ...


3

I have played around with those a bit and my results were mixed. Bollinger bands essentially show you the price relative to rolling window volatility. One interpretation is that if the current price leaves the Bollinger bands, a trend or movement emerges (of course depending on your time frame as with all technical indicators) in that direction. The ...


2

Let's approach the answer to your question from a pure trading and risk management perspective because looking at it from a mathematical standpoint nor quant standpoint does not yield you much here: 1) Bollinger bands are nothing else than standard deviation envelopes around the mean of past prices of the underlying. So, as far as simple probabilities go, ...


2

It sounds as if you would be interested in computing Relative Strength http://www.investopedia.com/terms/r/relativestrength.asp You could either measure it against a benchmark index such as the Dow 30, or compute your own index from your 50 stocks and measure each individual stock against the index. -Ralph Winters


2

Typically "average" lines are used to get rid of noise in the original data. It seems pretty logical to smooth intra week fluctuations when working with a year of data.


2

Scale and range are your biggest issues. If one input has values which range from e.g. 2300-3500, and another from 0 to 18, then the large scale of the first will swamp the other and provide greater informativity into your learning algorithm. Therefore, normalize into range [0,1] or mean-zero standardize - like you have already done. Be careful with ...


2

There is no right approach a priori. Try all approaches that make decent sense and pick the one that maximises a utility function on out of sample PnL and risk (or some similar decision rule).


2

John Bollinger, the developer of Bollinger Bands, provides descriptions of methods he suggests for using his bands on his website BBands.com. They can be found under Four Methods in the support area. Bollinger Bands are most effective when used with other indicators for confirmation, and are very powerful for mean reversion and for price breakouts. On the ...


1

You shouldn't recompute your EMA - just keep its old values, and apply your EMA formula to the last value to update.



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