0

I am getting confusing about Input data set . I am studying about Artificial Neural Network , my purpose is that I wanted to use the historical data (I have stock data of last 10 years ) to predict stock value in the future (for example 2015). So, where is my input? For example i have a Excel sheet data as [Column1-Date| Column2-High | Column3-low |Column4-opening|Column5-closing]

Sachin Chaudhari
  • 113
  • 2
  • 3
  • 9
  • 1
    With ANNs you have one input node for every feature type, plus a bias node which is always set to 1. If you have 5 features, then you have 5+1 input nodes. While I don't have any experience predicting the stock market, I don't think a simple ANN will do the job. Conventional ANNs don't understand time series data--they only see individual data points in space and try to create a function to predict them. They will not make any correlations between past and present. – eigenchris Jan 21 '15 at 16:22
  • You should spend some time to look for similar questions here on Stackoverflow. [This question](http://stackoverflow.com/questions/18670558/prediction-using-recurrent-neural-network-on-time-series-dataset/18692530) concerns the same problem. – jorgenkg Jan 22 '15 at 06:52
  • will feed forward neural network help in solving time series problems ? – Sachin Chaudhari Jan 22 '15 at 11:39

1 Answers1

0

By profession I am a quant and I am currently pursuing a masters degree in Computer Science. There are a many considerations when selecting financial input for a neural network including,

Select indicators which which are positively correlated to returns.

Indicators are independent variables which have predictive power on the dependent variable (stock returns). Common popular indicators include technical indicators derived from price and volume data, fundamental indicators about the underlying company or asset, and quantitative indicators such as descriptive statistics or even model parameters. If you have many indicators, you can narrow them down using correlation analysis, best subset, or principal component analysis.

Pre-process the indicators for use in Neural Networks

Neural networks work by connecting perceptrons together. Each perceptron contains an activation function e.g. the sigmoid function or tanh. Most activation functions have an active range. For the sigmoid function this is between -sqrt(3) and +sqrt(3). What this means is that you should normalize your data to within the active range and seriously consider removing outliers.

There are many other potential issues with using Neural Networks. I wrote an article a while back which identified ten issues, including the ones mentioned here. Feel free to check it out.

Stuart Gordon Reid
  • 51
  • 3