TidyTuesday: Modern Forecasting with Prophet and TidyModels

[@mattm9069]

Thanks! Always wanted to see how Prophet works.

[@deanmarris_1]

Very good explanations and efficient coding!

[@alanscott9258]

Great video. I'm a beginner in R so it was very fast for me but I can pause and think. Thank you

[@sebastianmullerbalcazar6229]

well done, your code is so clean.

[@knights_limit]

Appreciate your content and explanations! 😊

[@CohoCharlie]

Thanks man.

[@mcnamarachiwaye6359]

thank you Andrew

[@SideWalker_Canvas]

Hello dear. I am PhD student and i want to fixed upper and lower bound. i want to check forecasting when we want to apply policy and check does with these values of upper and lower bound then price will be change according to our expectation. I am using python

[@alexpreisler3647]

Thanks for the great demo, this really helped me out on a project and was an excellent introduction to prophet! I was wondering if you had any recommendations for learning more dplyr? I was definitely struggling to keep up with some of the group, nest, join, and mutate manipulations that you were using

[@AndrewCouch]

You can check out the "R for Data Science" textbook written by Hadley Wickham the founder/creator of the Tidyverse r4ds.had.co.nz/. I also have a video going over the basics of Dplyr kzbin.info/www/bejne/naOlZZSNaZZgg9k. The best advice I can give you is to practice all of the functions of dplyr on random datasets and see what functions you commonly use in an analysis. I personally have a lot of years of Tidyverse experience (3 years of academic experience and 2 years of professional experience) so don't feel like you should know everything, I certainly don't!

[@prod.kashkari3075]

I feel like I don’t know enough about the theory about time series to use these packages. Do you know what are some good books to gain some intuition before jumping into code?

[@AndrewCouch]

I recommend reading this if you want to learn about the basics of forecasting: otexts.com/fpp2/. There is also a paper about the development of prophet: peerj.com/preprints/3190/

[@CaribouDataScience]

Why include history =True?