AI in finance – How we predicted NASDAQ Stocks
Since the genesis of AI (Artificial Intelligence), we have come a long way by using it to affect a wide range of applications. We use AI in various sectors like Finance, Education, Healthcare etc.
One such sector where humans have always thought of using Machine Learning and AI is for stock prediction. Countless people trade stocks every day. Many experts analyze the stocks to forecast it by minimizing risk. Predicting the values of these stocks beforehand would obviously be a great advantage. Before discussing our approach of using Machine Learning for predicting stocks, let us try to understand the patterns of stock markets first.
What are the stock prices actually dependent on?
What did we want to build?
Well, now it is clear that stock market prediction is quite a complex job and even if we do build a model that can predict a market stock, it’s highly unlikely that it will be accurate enough to try your money on. So what can we do is rather than predicting one stock of the market, we can predict all stocks or at least multiple stocks (including most important ones, of course) of the market.
How would this be helpful?
As I said earlier, the important part is not how much a stock will go up, its how a stock will go up in comparison with other stocks, especially the one you are planning to put your money on. That is what we want to do. If you are planning to put money in say, either Google or Apple, the prediction will help you decide which stock to put your money in.
Our Approach –
1. Adding data and Updating data :
2. Initial Training:
3. Incremental Training:
Contributed by: Saurabh Lodha