I'm a first-year PhD student who is quite new to time series forecasting, my research topic is to implement deep learning algorithm to forecast time series data. I've already reviewed some recent papers about this topic, by considering many papers integrate deep learning with statistical approach, I'm thinking if there are some good books which can help me get start in this field. And I will be very appreciate if someone can give me some suggestions of the learning roadmap.
https://www.oreilly.com/library/view/practical-time-series/9781492041641/ - This book will be a great beginner's course on the topic and you can refer back to it time to time. It's even part of a free trial option on o'reilly.
This will give the basic introduction. You can then go to MIT OCW for a pure statistical approach or solve questions on Kaggle with your specific domain in mind. https://arxiv.org/pdf/2004.13408.pdf This review explains the history in short and you can also approach one of the future directions mentioned here. There is a course on coursera as well which deals with time-series forecasting, it will be like an abridged version of the book.
Finally once you have spent some time, you can focus on a single aspect and try to search for a specific course in that topic itself.
I think "Forecasting: Principles and Practice" is one of the best resources in this field. You can find it as a free online ebook here:
https://otexts.com/fpp2/
Hi Shuyang, you can join the courses to learn basic concepts.
https://www.coursera.org/learn/practical-time-series-analysis
https://www.coursera.org/learn/tensorflow-sequences-time-series-and-prediction
It will be helpful to get the confidence about the topic. You can refer papers related to your topic and list out the strength and weakness of methods applied in the research papers.
- Badges
- Science topic
- Similar topics
- Computer Science
- Research Topics