PREDICT STOCK PRICES IN REAL TIME WITH FACEBOOK’S PREDICTIVE MODEL USING MACHINE LEARNING

Author (s) : C.Prabhadevi and S.Jayanthi

---

Abstract :
Predicting stock market performance is one of the most difficult things to do. Forecasting involves so many factors - physical factors vs. psychological, rational vs. irrational behavior, etc. All these factors together make the stock price volatile and very difficult to predict with great accuracy. Can machine learning be used as a game changer in this industry? Using features like the latest announcements about an organization, their quarterly earnings results, etc., machine learning techniques can reveal patterns and insights we haven't seen before and can be used to make unerringly accurate predictions. “The resource is simply one of the most common things that we use in our daily life. For example, Averaging scores to determine overall performance or finding the average temperature over the past few days to get an idea of ​​today's temperature are all routine tasks we do regularly. So it's a good starting point to use our dataset to make predictions. The predicted closing price for each day is the average of the previously observed set of values. Instead of a simple average, we use a moving average technique that uses the most recent series of values ​​for each forecast. In other words, in the later stage of caching, the predicted values ​​are considered, while the oldest observed value is removed from the set. Here is a simple diagram to help you understand it more clearly An existing project that uses the simplest machine learning algorithm that can be applied to this data is linear regression. A linear regression model provides an equation that determines the relationship between the independent variables and the dependent variable. We use Face book Prophet, an algorithm developed by Facebook’s Core Data Science team. It is used in time series forecasting. It is widely used when seasonal effects are possible. Time series forecasting is widely used in stock price forecasting. In this article. I will walk you through applying the Facebook Prophet model to predict Google stock prices.

Keywords : machine learning, linear regression,Core Data Science

Conference Date : 30/08/2019

Pages : 14-15

Paper ID : 19014