R TUTORIAL: Forecasting Airline Travel COVID19 | NEW Modeltime Features

I spent the last 6-months adding NEW #TimeSeries #Forecast tools to the #Modeltime Forecasting Ecosystem in R. Today, I show of 4 of the new features with a FULL R Tutorial. We analyze a critical business problem: forecasting airline travel and domestic passenger load for four major airlines.

We use the following new features in Modeltime to build forecasts for each airline:

1. NEW Modeltime GluonTS & Torch Deep Learning Algorithms
2. NEW Workflow By ID Features
3. NEW Hyper Parameter Tuning & Parallel Processing for Machine Learning
4. NEW Global Baseline Models

📰 WANT THE CODE SHOWN TODAY?
Join Learning Labs PRO: https://university.business-science.i...

📈 NEED TO BECOME AN EXPERT IN TIME SERIES?
Take my time series course: https://university.business-science.i...

🏁 JUST STARTING OUT? NEED TO BECOME AN EXPERT FAST?
Take my 5-Course R-Track: https://university.business-science.i...

Table of Contents
00:00 New Forecasting Tools: Modeltime & Modeltime GluonTS
00:49 Goal: Forecast Airline Passenger Traffic with COVID19 Impact
01:29 About Learning Labs PRO Program
03:03 Business Problem: Airline Passenger Forecasting
05:06 Modeltime Ecosystem: Growing System of Forecasting Tools
10:33 Lots of Models Available in Modeltime
15:48 New Forecasting Tools We'll Use Today
22:52 Full Code Tutorial (Starts Here)
23:45 Project Setup and GluonTS Installation
26:16 Libraries (tidymodels, workflowsets, modeltime, modeltime.gluonts)
27:38 Data Import
29:04 Clean Data
32:55 1.0 NEW GluonTS Deep Learning Models
39:24 2.0 NEW Workflow By ID Features
45:34 3.0 NEW Hyper Parameter Tuning & Parallel Processing
55:12 4.0 NEW Global Baseline Models
1:05:09 Final Forecast
1:07:29 About the Time Series Course
1:10:02 About the R-Track & Time Series
1:14:53 Student Transformations
1:16:50 Q&A