Do you have any handy references emphasizing data cleaning and preprocessing for NN? What NN does it replaces hundreds to millions of conditional statements with a grid of activation functions that converge to a single result which depends on having enough distance in statistical terms between groups. There is nothing complicated about it except for the complexity of the design.
There are many books and articles published regarding NN applied to trading in the last 20 - 30 years, some with more concrete designs being shown together with "promising" results. None of them proved to be working but on the other hand no one is going to offer successful design to the masses for obvious reason.
you are either a hopeless case or are bringing this data cleansing issue up again and again and again for some, to me, obscure purpose. I am tired and don't think I will stick around longer in this thread. Best wishes to you.
Before using NN or machine learning read this paper: http://rnm.simon.rochester.edu/research/MSES.pdf
Absolutely not. You are way off the mark with comparing AlphaGo to trading and completely wrong that you can use raw data. Board games or video game play is an unsupervised machine learning problem, trading is a supervised machine learning problem. A win in a board game is the result of a series of moves, trading profits are the result of average profit factor for every trade. Completely different problems. In addition, historically any potential entry can be tagged as a trade loss or gain, making this a supervised problem. Treating it as unsupervised is simply the wrong approach. In addition, trading has considerable uncertainty, where board games have only one outcome. A more similar problems are computer vision, eg identifying the object in a picture, or search engines, or natural language processing. The best example is the Viola-Jones algorithm for face recognition used on smart-phones. These algorithims are completely dependent on the "features", or mathematical conversion of the raw data into something the computer can understand and then convert into a probability. The breakthroughs in these fields come from the development of better features, and what algorithm (neural network, support vector machine etc) does not make much difference. Also, computers can outplay a person in chess, but a two year old can describe images better than a computer. A person could also provide better answers than a search engine. What good answers the search engine does give such as local results based on knowing your location or knowing a number is a fed-ex tracking are a hard wired features. And look at the recent crashes from car autopilots. Similarly, good trading algos can be built but it is by no means as trivial as and requires considerable feature creation. No, you can't just feed raw price data into a ML algorithm. That is absurd.
Can you be more specific? I am trying to find some algos in neural network to adjust my indicators. Thanks ! As I became enthusiast to neural network quite recently, I am wondering whether R is a ideal software to implement neural net works? As far as I know, R is used by many professional statisticians and could run the Monte Carlo too. If the answer is yes, I don't have to waste my time on other platforms like Matlabs, right?
Can not agree more with you. Market transaction is more complex than playing a Go game. It is highly suspicious whether ML can recognize any statistically significant model of the market. So, it seems to be simple minded to apply ML directly to generate buying and selling signals. So, what about the idea that we use ML to adjust the parameters of indicators or other similar way ?
R has a much larger user base than Matlab, and therefore also much more libraries for neural nets and other algos. For deep learning, look into DARCH, DEEPNET, and DNN.
All, I've been reading up and I want to post a correction to this post. It appears that Viola-jones has long been outdated and now convolutional neural networks are used for image detection. This does not require feature engineering. Also, most board games (other than Go) are essentially search algorithms. I'm thinking now that I will pursue LSTM (a type of CONVNET with short term memory used for speech recognition) with reinforcement learning (TD(lambda) or such) for the trading algo portion could actually work with minimal feature engineering as long as you build the LSTM large enough.