Weekly Challenges

Solve the challenge, share your solution and summit the ranks of our Community!

Also available in | Français | Português | Español | 日本語
IDEAS WANTED

Want to get involved? We're always looking for ideas and content for Weekly Challenges.

SUBMIT YOUR IDEA

Challenge #24: ARIMA Time Series

mikeprice17
8 - Asteroid
 
Pang_Hee_Choy
12 - Quasar
 
mpennington
11 - Bolide

Solution attached.

 

Spoiler
MP24.png
SamDesk
11 - Bolide
Spoiler
Annotation 2020-03-16 163150.png

Hey @AYXAcademy, think the solution needs updating as the workflow produces a different output from the stored output values. Probably an update in the R package at some point as caused this.

 

Sam 🙂

AngelosPachis
16 - Nebula

Challenge #24 solved

avinashsunchu
8 - Asteroid
Spoiler
Solution.JPG

kaianderson
8 - Asteroid

Here's my solution!

atcodedog05
22 - Nova
22 - Nova

On a spree to binge complete weekly challenges

 

Third times the charm. I got my numbers to finally match on the third time 🙂

 

Spoiler
This is my solution

atcodedog05_0-1585209168403.png

 

AndrewS
11 - Bolide

#24-same variances on forecasts.

steven4320555
8 - Asteroid

As previous predictive analysis exercise, the challenge is good to get an understanding on how to connect the flow to make a general prediction. But the default output is not that convincing. Ideally, everyone should have difference predictions depending on how people model the data, but exercise is exercise 🙂

Spoiler
steven4320555_0-1586780807666.pngsteven4320555_1-1586780845365.png

The default fitted model has a time-lag, and tend to give higher estimations for starts of the year. I would love to estimate something around 100, based on previous two years' data.

steven4320555_2-1586781106373.png

I tried to play with the adjustable parameters, but came acorss some errors.

steven4320555_3-1586781613313.png

It is easy to build the model, but will take some time to understand how the model work. Will time sometime read the recommended reference material https://otexts.com/fpp2/