Hello Everyone
,
I'm focusing on a project right now that uses time series analysis to anticipate stock values, and I would really appreciate your thoughts and recommendations on the best approaches and strategies to take.
To briefly explain, the goal of my project is to create a prediction model that, using past data, can anticipate stock prices with accuracy. I've compiled a sizable dataset with daily stock values for the previous ten years. Though I am a novice to time series analysis, I am familiar with fundamental statistical approaches and have some expertise with machine learning.
I have the following specific queries:- Which historical data models are most suited for stock price prediction?
I'm not sure which of the ARIMA, SARIMA, which and LSTM models to use or if there's any more models I should take into account despite having read about them all. - How should I respond to data trends and seasonality? Seasonal patterns and trends over time are frequently seen in stock values. Which methods work best for locating and adding these minitab elements to my model?
- Which typical mistakes should one avoid when handling time series data? I want to make sure my model is trustworthy and strong. Are there any typical errors or difficulties that I ought to be mindful of?
- What resources or tutorials would you suggest for someone who wants to learn more about longitudinal analysis? We would be grateful for any books, distance learning programmes, or particular articles that you noticed to be beneficial.
I also followed this:
https://ieeexplore.ieee.org/document/9776830Thank you in advance.
