Applied Time Series Analysis in Python (Udemy.com)

Yes. It is so helpful to watch you iterate over the possible parameters and capture the lowest AIC. It seems very practical. I also enjoy the theoretical coverage of this material.

I enjoyed how simple the structure was regarding each topic and it's implementation. Having the repetitiveness of applying the different models but with similar code is good in terms of practice and rigour, but that time could have been used to dive into more detail on theory. Even though a lot of knowledge is assumed, skimming over deep learning is not helpful when deciding what type of framework to use by intuition. In terms of code improvement, maybe making the classes that you wrote to be a bit more abstract could help them become more useable for different projects.

Things were going decently well until Deep Learning Project came in. The guy did not explain anything about the Window Generator class he built. It seemed like he copied from some other code and just typed it during recording. After that the different types of model fitting was still ok. But unless we understand how the WindowGenerator is working, it becomes very difficult to understand the rest of it. Also a bit of focus on the installation process and error handling would have been better in the Prophet project.

First of all, thank you Marco for creating this useful content. Overall the course was a good introduction into time series forecasting but I believe a lot of issues still exists. 1. Most of the theory was skipped which I feel is important to even implement any algorithm effectively 2. I personally felt that some of the code was not correct, especially when Marco tried to get the forecasts 3. Deep learning module is extremely high level. Even though, I have some prior experience using DL for time series, it was hard for me to keep up. 4. Lastly. I think this course was posted a while back and hence the Q/A forum is not active. Therefore, you can find it difficult to get answers to your queries.

I feel the course was presented with excellent practical material but it was missing on two things: - a little more theory, - more thorough interpretation of the findings Additionally, it could be improved by inserting relevant definitions ahead of the sections that use them first time; like the definition of AIC for example which comes a lot later after it is already been used.

This course is excellent on the overview of Time Series Analysis. Even though I'm new to Python, the projects are also reasonably easy to follow. Thank You

What is great about this course is that it builds up slowly, explaining very clearly the basic concepts and then it keeps adding complexity - this makes it very easy to learn! I also appreciate the fact that it also teaches theory through practice - explaining it as it goes. What I don't like about the course is how much time it spends with plotting - this adds no value but fills up a lot of space -, that it is very repetitive, at least for the ARIMA to VARMAX part, and that it gives no assignments for us to do individually and compare to the teacher's results - I would feel quite stupid just re-writing all his code and convincing myself it's a code exercise.

This course, is really what I searched for. A intermediate course for timeseries. Its a good way to start a course with a review of the basics, because afterwards everyone is on the recommended/needed level. I would do an other course from this instructor immediatly. Nice job!

As a professional data scientist, I was looking for a way to learn time series because I need it in my job. I had very limited knowledge on this subject. With this course, I learned everything I needed and it helped me a lot to succeed in my project. Marco is a great teacher! Thank you! I recommend it to everyone who want to learn about time series!

I passed until Deep Learning part . My opinion first part is an immediate recipe , a hands on aproach .If you have basic (python + TS flavour) you can start practicing with good results. Pick 10 well known TS datasets and go get it..Repeat the procedures and that it ..at least that's my plan . Thank you Marco for opening the path in a simple way.