If you do not care about how the neural network works then it is quite probable that you will not find them useful. As has been mentioned the activity of "neural networks" (though they have little neural in them) is interpolation. The value however of neural networks or other machine learning techniques is dimensionality reduction, and potentially the ability to find moderately low dimensional projections of complex, high dimensional data, but ones which would have eluded humans on their own. Other, faster techniques may be more suitable for initial exploration of complex data for the purpose of extracting the useful subspaces, and then a neural network can provide a somewhat higher quality fit than the simpler, but faster techniques used as approximations in the dimensionality reduction. Another value is that the interpolation in the moderate dimensional space can be controlled with regularization (as can other statistical techniques) so that it is 'soft' (low derivatives) and hence vary smoothly between known data. If you don't understand this message, then you ought to stick to using the neural network which comes standard.
I was referring to the NN that are sold "ready to use". That was the case of the link posted earlier.
And again you are not right. You might read this article quite superficially. The authors know the main obstacle in implementation of NN. That is complexity to understand how NN works. That's why they propose to train Artificial Foreteller their selves based on user's tick data. In that sense you are right - it is sent to end user as "ready to use". Moreover, user can order a system that will use user's set of indicators. And I think that it is quite effective approach when user will use this system only as a tool that helps him to be more productive. I do not think that you know exactly how CPU works in your PC. But that doesn't prevent you from using it.
http://www.calsci.com/S&P500.html This is the NN that i have. LBS Capital appears to use this NN to trade in almost all their programs. Perhaps it would be useful to check their performance. BTW: The brain has biological neurons that fire at about 100Hz, and there are about 100 billion of them, with 10,000 synapses for every neuron, all being modified in real-time. How big the giant supercomputer would have to be to match the human brain? Perhaps tens of peta-flops? That's probably why my brain achieved something that i couldn't achieve with a NN. My brain is more powerful, even with the limited intelligence that i have, and probably also because i don't know how to use a NN to be successful.
From article in:Stanford Report, March 20, 2006 After earning his doctorate from Caltech in 1997, Boahen became a professor in the Department of Bioengineering at the University of Pennsylvania. During his eight years at Penn, he developed a silicon retina that was able to process images in the same manner as a living retina. He confirmed the results by comparing the electrical signals from his silicon retina to the electrical signals produced by a salamander eye while the two retinas were looking at the same image. In January, Boahen came to Stanford, where he will turn his attention to studying how learning and memory work in the human brain. "What we're trying to do nowâwe've come up with ways of modeling neurons and synapsesâis to build chips with something like 100,000 neurons on [them] and then build a multiple-chip network that gets up to about 1 million neurons," Boahen said. "With a network of that size, you can model what the different cortical areas are doing and how they are talking to each other." His goal is to eventually create a silicon computer that works as efficiently as the human brain. According to Boahen, the brain is capable of performing 10 quadrillion (that's 1016) "calculations," or synaptic events, per second using only 10 watts of power. At this rate, he says, a computer as powerful as the human brain would require 1 gigawatt of power. Question: Howcan a NN that is sold today for about 1000$ run on a home PC and produce results of good quality knowing that, accordingly to experts, we need huge computerpower (that we don't have right now) to imitate the brain? There is still a long way to go.
When tomorrow every trader owns a computer with that kind of capability, probably the oevrall profit/ loss results for individual traders would be still about the same. Are we expecting today's losers would become winners tomorrow because of that? If yes, then there would be no any losers in trading ?
Indeed, except for those who will not have a supercomputer. They will already lose before they get started. The battle with continue on a higher technological level. Like in every sport. But there will always have to be winners and losers.
You've made a huge leap in this post and the prior by assuming that neural nets need to be able to imitate the human brain in order to be of use or be superior for certain tasks. For example, can your human brain monitor 623 indicators simultaneously and recognize patterns in those indicators real-time? As I pointed out earlier, there are many tasks that can be achieved more efficiently by computers compared to a human brain. I thought that was stating the obvious! Generalization versus specialization. You seem to be sliding towards a logical fallacy: - Neural nets are not as powerful as human brains. - Therefore neural nets are not as good as human brains for pattern recognition/signal generation in trading systems. - Excel is not as powerful as a human brain. - Therefore Excel cannot peform multiplication as well as a human brain. Good luck. MoMoney.
Interesting thread... I wasted 2 years on Neural Nets. I purchased a lot of software including packages from BioComp, Ward and Braincell. The best package that I used was SNNS. It was best because it had the best documentation, user interface and examples, runs on Linux and Windows, and was free. The commercial packages have better UIs for newbes but at the end of the day you will probably be using scripts anyway so the UI does not help. I trained the system daily using many different markets and was looking for an intermarket edge. Some thoughts from that time: Coordinating data is a bitch. One data-fart can cost a lot of money and be very hard to find. Often I would not know that a data-fart exists until after I would lose a lot of money and go looking for the cause. There are tools that can help but not much. Unless you have to time to write real data validation routines that can do more than identify when a time series has shifted 1 or more bars or the data is way outside the range of the last 20 bars.... Maybe validating data against multiple sources would work. I was able to get the same or better results using less and less data as I learned more about the problem. Throwing tons of data at a NN and expecting it to sort everything out does not work (IMHO)... More on data. What is the best way to handle 'market shocks' with a NN? I don't know. Market shocks in historical data messes up your results. If the shock was a scheduled report maybe you can plan to handle that day differently. You could clean your data to remove market shock spikes, but what will your system do when it faces that unscheduled market shock tomorrow??? One way many improve the NN results is to 'train every day'. Fine. The system signals a LONG, the next day the market goes south, the system says BUY MORE, market goes faster south BUY LOTS MORE, you retrain the nets that night and the system says, 'SOLD 3 DAYS ago, SELL MORE!' WTF!!!! What about 'walk forward testing'. Some vendors claim 'walk forward testing' is not 'curve fitting' because even newbees know curve fitting can be dangerous. Curve fitting can give the user false confidence that their curve fitting results during the sample data period did not destroy their account and then find that even routine moderate volatility days can completely eliminate profit goals reached during testing. The vendors claiming walk forward testing is not curve fitting are being dishonest to the point of committing fraud in my opinion. The curve fitting and walk forward testing assume that tomorrow there will be a buyer for every seller and a seller for every buyer. Not always the case, 9/11, Oct 1987, and LTCM are good examples. The NN is a trend following system and needs wider stops than most trend systems because of the noise in the forecast. Even a good system in backtesting will have 5 losing trades in a row. If you are trying to forecast a price 2 days in the future or 6 days in the future you will probably need a 1.2 or a 1.5 * AvgTrueRange of 2 or 6 days, this plus 5 losing trades in a row starts to get expensive. Managing risk is very difficult to do as the nets are always changing. Every time you re-train the system; the system you will use tomorrow is new. Think about that statement, tomorrow's NN is NEW. Your results from the past 3 months have nothing to do with tomorrow's NET retrained last night. With a state machine or a GeneticAlgo you can track down the specific trigger for an entry or an exit. With a multilayer NN you can look at the layer weights and inputs but it is hard (impossible for anything except the most simple net) to find the trigger that gets into into/out of a trade. Just how much do you trust your net? One guy on the web, around 2000, started a site and published his real time results for 20 markets. He did great for 6 months, people were subscribing, sending money. Looked great. Then the market turned. All of his NNs were on the wrong side of the market and stayed wrong for about 3 weeks. It was ugly. He held fast to his NN until he reached 50% drawdown and then closed the site. Each day he would publish a 'holding fast WE believe our technology' letter to the web site. As long as the behavior of the market says either volatile or within a range the nets are fine. However, when the markets turns from a trend to a range period or from range to volatile, look out baby!!!! How can you be confident when you have had 4 losing trades in a row and not wonder if you are moving to a market fundamentally different than the one used to train data??? You cannot. IMHO I started to scale back my NN project and just use NN to replace indicators such as RSI. In effect making the NN a better filter. Not predicting the future as much as reducing noise. This is a good application of NNs. However, there are better filters that will do the same without the overhead such as T3Average and the Jurik Research products. (You can find alternatives to Jurik's stuff on Tradestations forum. The Tradestation stuff is free. However, if you want to play with NNs you probably have a fat software budget so spend away. Jurik's indicators are good and cheap for what they do.) Can a system be profitable and not re-trained 'often'? The fact that all NN vendors retrain and now have tools to choose from 1 out of many alternative systems is proof that they do not work. "They can demo, but they cannot trade." Big difference. The good news is that computers are cheaper now and much faster. Databases are much better. And best of all, if you spend 2 years developing a NN you can learn enough about the market to reach a point where you can say, "Hey, I can do this good with a price action or KISS system and enjoy my evenings and weekends!" Only then will you start to focus on money management, position size, and price action and maybe make some money. Maybe. Lastly, the biggest advocates of NNs have something to sell, or are in denial (halfway into the woods). Don't believe any 'historical testing' of NN as it is fake. I would only believe system results that were gained trading real dollars. It doesn't have to be large size, 100 shares of anything would be fine. Trading micro-pips at oanda is fine but it has to be something real. You won't find any vendor providing it or names of anyone who you can talk to not vested with the software provider. Somebody prove me wrong and post the result here please or where they can be found. I still have all of my work on the cold linux boxes over in the corner. Give me a reason to attach them to a monitor!!! Sorry for the Hershey length post!!!! Good luck!