Plotting Time Series in R: A COVID-19 Example

Here I walk through an example showing how I'd plot time-series data in R using the ggplot2 package. In this plot, we look at coronavirus COVID-19 **new** daily cases in the United States and Australia.

🎉 *Enjoyed this video?* Leave a comment below to share what you liked the most!

0:00 Intro
0:15 Download the data
1:00 Read the data into R
1:40 Re-arrange the data to prepare it for plotting
5:18 Plot United States cumulative cases
6:00 Plot United States new daily cases
7:13 "Beautify" the plot (add labels etc.)
9:55 Multiple lines (Australia & US new daily cases)
11:05 Split facets using facet_wrap()
12:15 Remove the legend

[tomhenry-datasciencewithr]

😃 Comment below to share your own ideas for time-series plots!

[choma]

This is mind-blowing! I would have never dreamed to get to a so lovely code.

[williammendieta]

The best of the video is the data manipulation! Very helpfull!

[ahmedJaber]

thank you for this Tom. Wonder how you would complete a timeseries data that misses a number of dates . e.g. 2022 data that doesn't have 1-20 Aug days

[AsrafunNaharAnonna]

Neat and clean elaboration. Helped me in my project work. Such a life-saving tutorial. Thank you so much. :D

[justinarends]

New to R and am self-teaching using the R for Data Science book. Your walkthrough was great and easy to follow. Will definitely be watching more of your content!

[muhammedhadedy]

Excellent tutorial. I wish if you make more of these videos.

[jorgersoldevilairizarry]

Thank you so much for this. I am new to R and spent an entire day trying to build a similar program but was unsuccessful. This video helped me tremendously in finishing my project.

[susanbenson]

Great video. Clear and concise. Thanks.

[fatimaejaz]

Can you also explain if we set different time intervals as our y axis for the graph

[JobGathirufisher]

this was useful, clear and loaded with tips.

[jaimemadrid]

Hello Tom, thanks for your videos. I'm traying to display the last value (new cases) of each line as label from this example, how can I do that?

[farzadraeisi]

great and useful. I have a big dataset with 500, 000 observation ( some categorical variables (like department, disease, ...) and several datetime which is related to a hospital (Admission date-time, Doctor Note date-time, discharge date-time, ... ). how can I make some discrete time period based on hours, ex: morning (from 6-10 am) to produce informative plots?

[ahmedalnajar]

Thank you so much. I wonder if you could change the Y-axis scale instead of separating the curves, what would you use?

[crazyghost]

Thanks a lot. Your tutorial helped a lot. For your information, your provided code link doesn't work.

[freddiebrown]

Really nice walkthrough. A couple of cool tips I’d not used before (lag and scales::comma). I would love to see some statistical techniques for forecasting and comparing time series

[reem]

Hello, wonder how to plot the number of cases per year, if I have a date of diagnosis (variable) for each patient(variable)?
thanks

[swingingarashi]

Thank you for this useful tutorial! I also tried to visualize the covid data in my country but i couldnt replicate yours because everytime i try to plot the date by date on the x axis it wont come out as months like you did, but instead it would be something like (the beginning date) - (kind of middle date) - (last date). I also used lubridate, ggplot, and tidyverse so what should i do to make an output like yours?

[MarkIreland_carrierband]

Do anything similar in Shiny? How about the rate of new cases/population?

[diablo]

im getting error saying "n_confirmed obj is not found" what to do?