How function approximation predicts the stock market?

During history of human life, people have attempted to find new ways in order to response their demands. According to the nature and issues of different problems, they have created new approaches or optimize previous ones to increase efficiency and gain much more benefits from them. Following this trend, seeking for new ways to help people in their newly created challenges is necessary. In the following paragraphs a new method for predicting variations in capital market will be introduced.

Being successful in capital market and gaining more money was one of important issues in human life. By advancing in technology, capital market relations getting more and more complicated and consequently, modelling them as sets of parameters also getting more and more difficult. On the other hand, knowing changes and variations occurred in the world and their effects on market are very important and valuable as they would give people power and wealthy.

Omid Saremi et al. presented a new approach of predicting ongoing events and variations based on previous situations by introducing of revised classifier systems. This method of classification generates an approximate function based on historical data. Then, the system is able to predict the future trend based on the function. There are many applications for this type of algorithms; but as we focus on business in following article, this aspect of the paper would be developed in continue.Stock-Market.jpg

Photo Credit

One of strategic places for investing in all countries is their stock exchanges. The first subtle point that should be aware of is when and where you invest your money. Various events and situations change trend of variations and as an investor, this could get you wealthy or bring you bankruptcy. Assume that you want to invest in stock exchange; at first you should collect data from previous variations and the events that made those changes (this is important to know the market before entering this field). Although many parameters influenced on stocks variations, you choose important ones as they are almost good candidates for expressing changes. For instance, we want to study midterm and longtime influences of oil price on General Motors stocks. By gathering General Motors stocks information as well as the parallel oil price and provide them to this algorithm, it trains to predict changes for any new situation. When training procedure finishes successfully, you enter a state in which the stock price is X $ and the oil price fall about Y% in previous week. By providing the state to the algorithm, it maps the training data on this situation and predicts future changes for the stocks. This information would be very helpful for investors whether to sell, buy or just keep their stocks to gain more benefits or avoid losing money.earning-losing-money-14881323.jpg

In conclusion, seeking novel ways to apply on new issues is necessary for human beings. In this regard, using new approaches for helping people to make right decision in different conditions at various fields like business is useful and beneficial. To response this demand, an algorithm for predicting ongoing changes based on previous events that is capable of using in business was designed and introduced. The algorithm could train according to a set of parameters and finally approximates the changes for provided new situation. The information that is absolutely useful for investors and businessmen, and in general, for decision support system of any company. What do you think about other application of function approximation?

 

Advertisements

Author: Amin Sabzehzar

MBA student Mechanical Engineer University of Nevada, Reno

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s