WEEKLY DIGEST 2015, February 01 - 08 for Neural Networks in Trading: Do NN need “compound” features, and How do I normalize data for stock prediction?
mql5 blogs
"A neural network is essentially a system of programs and data structures that approximates the operation of the human brain. A neural network consists of a large number of processors operating in parallel, each with its own sphere of knowledge and access to its own databank."
============
Do Neural Networks need “compound” features?"Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It's 100% free, no registration required."
============
How
do I normalize quantitative data, such as Open, High, Low, Close,
Volume, Adj Volume, of NSE stock market for stock prediction using
artificial neural network?
Q&A on Quora
"Some
have tried not using price data at all, and instead using indicators
based on that data. For example, you could have an input that switches
between 0, 0.5, and 1.0 depending on whether the market's close is
below, within, or above its Bollinger band. Oscillators with a fixed
range are also popular for ANN inputs precisely because they are so easy
to normalize.
Also, remember that you have to do something with
the ANN output, which is just as difficult as figuring out what the ANN
inputs should be. What sort of training algorithm are you going to use?
If you have the output correspond to predicted price movements, then you
still need to develop a trading strategy based on that. If you have the
ANN's output actually be a trading signal, you also need to figure out
how you're going to train it.".
============
TRADERS’ TIPS
"For this month’s Traders’ Tips, the focus is Dave Cline’s article in
this issue, “Candlesticks, Condensed.” Here, we present the February
2015 Traders’ Tips code with possible implementations in various
software.".
