A question about forecasting

Hi, there:



 I noticed that time series can be used to forecast the prices of securities… but I wonder… if I have a historical data of a stock… I have OCLHV data… is it possible to forecast the low of the next period using time series?? 



I may want to to this in order to buy dips… I wonder if any other people have this idea…

Hi Ghery



This question has a very broad answer. First of all, the OHLCV data that you have is indeed time series data. So what you are trying to achieve is basically predict the low of the next candle. 



It is not impossible. But it is not easy either. Many machine learning techniques ranging from simple linear regression to advanced deep learning models involve time series forecasting as one of the applications in the financial markets. 



To get started with time series though I suggest you to go through this blog: All About Time Series: Analysis and Forecasting



Hope this helps.

Thanks a lot… I heard that the model commonly used to  forecast the price of stocks is the ARIMA time series model… can that be used to forecast the low of the next candle  maybe???



If it can be used as such…  how do you determine the parameters of the arima model??





 

Hi Ghery



People generally use ARIMA to forecast the close. I think the same methodology can also be applied for predicting the low. 



For implementation and further details I am sharing two articles below:



1. Stock market forecasting using Time Series analysis With ARIMA model

2. Time-Series Forecasting: Predicting Stock Prices Using An ARIMA Model



Hope this helps.

thanks a lot… I really appreciate it…