Sir, I have always been thrilled by your R techniques and the way of your explanation. Your videos have caused me to switch to R [and of course RStudio] from Python [Jupyter] literally captivated me in enjoying analytics using R. I am working in Aerospace Industry. While the organization leverages a couple of premium visualization tools even to analyze exploratorily, I, after having got lightning stuck by your amazing videos, have been using ggplot extensively along with my basic statistical knowledge and of course teaching my colleagues. Thank you, @Riffomonas. Please keep posting more videos to enlighten some of thirsty analysts like me.
Another great episode. What would be very cool to do is take this data and try and forecast precipitation using an arima time series model. Thats prob going to be a few videos.
Thank you very much!! One question, do you know how to have different date_formats depending on the date? I mean: you have a timeseries which is hourly but I want to show in the x-axis for example: "17/05/2023", "06:00", "12:00", "18:00", "21:00", "18/05/2023", "06:00"... and so on
Nice presentation. It takes quite a bit of work to tidy up the plot in the end. As I have mentioned earlier I have data from another station here in the upper midwest. However, the station stopped operating in 2005 so I used 2004 as my reference year. The data streches back to 1984 and what I see is actually is that the first six month are if anything drier. Note we are dealing with precipation but have happens if we include snow for the winter months? I did mention in a previous message that I has missing tmax data in the early 1900 and it looked quite serious. However, when I inspected the original data with the skimr and vizdat packages I found in fact that only 1 year was missing all the data - not just tmax. It might be a good idea to check the data for "missingness" before diving into extended analysis. That was the lesson I learned. Finally, some on mention that a video/exercise on time series analysis might be fun. I agree. By the way there is an interesting pacakge for dealing with time in R called padr. It might be worth looking at. Thanks.
They convert snow into precipitation equivalents by melting it. Make sure you saw my note in the video that what I initially had removed all of the February data
My local data has other years with big gaps in it. If I wanted to check for complete years (every month represented at least 1 time in the year) I would use a window function in sql -- do you know a good way to query this in R? The closest thing I can think of is to use some sort of summarize len(year) and then make an assumption that I should have at least 300 datapoints/year, otherwise drop it
Something I forgot to mention in the video was that at the end I ran renv::snapshot() to update my renv project with ggtext as a dependency.
This tutorial is not only the best video for know how to use lubridate and ggplot, its a masterpice of how to creat great plots-
Thanks Luis! That is very kind of you to say 🤓 thanks for watching
Sir, I have always been thrilled by your R techniques and the way of your explanation. Your videos have caused me to switch to R [and of course RStudio] from Python [Jupyter] literally captivated me in enjoying analytics using R. I am working in Aerospace Industry. While the organization leverages a couple of premium visualization tools even to analyze exploratorily, I, after having got lightning stuck by your amazing videos, have been using ggplot extensively along with my basic statistical knowledge and of course teaching my colleagues. Thank you, @Riffomonas. Please keep posting more videos to enlighten some of thirsty analysts like me.
Thanks for your very kind comment!
Thank you so much for your amazing videos. I hope our comments help the algorithm put your videos on top!
Extra ordinary... Great job!
Thanks Ali!
Another great episode. What would be very cool to do is take this data and try and forecast precipitation using an arima time series model. Thats prob going to be a few videos.
Hah! I might need to produce another video… 🤓
@@Riffomonas happy to jump in and help hahaha
Fantastic. Thanks for your content!
You are the best! It was very fun and useful, as always!
Thanks!
Thank you very much!! One question, do you know how to have different date_formats depending on the date? I mean: you have a timeseries which is hourly but I want to show in the x-axis for example: "17/05/2023", "06:00", "12:00", "18:00", "21:00", "18/05/2023", "06:00"... and so on
Nice presentation. It takes quite a bit of work to tidy up the plot in the end. As I have mentioned earlier I have data from another station here in the upper midwest. However, the station stopped operating in 2005 so I used 2004 as my reference year. The data streches back to 1984 and what I see is actually is that the first six month are if anything drier. Note we are dealing with precipation but have happens if we include snow for the winter months?
I did mention in a previous message that I has missing tmax data in the early 1900 and it looked quite serious. However, when I inspected the original data with the skimr and vizdat packages I found in fact that only 1 year was missing all the data - not just tmax. It might be a good idea to check the data for "missingness" before diving into extended analysis. That was the lesson I learned.
Finally, some on mention that a video/exercise on time series analysis might be fun. I agree. By the way there is an interesting pacakge for dealing with time in R called padr. It might be worth looking at.
Thanks.
They convert snow into precipitation equivalents by melting it. Make sure you saw my note in the video that what I initially had removed all of the February data
Makes sense. The term precipitation is the broad term.
Everything is cool! Thanks!
🤓
Thank you!
You’re welcome! 🤓
My local data has other years with big gaps in it. If I wanted to check for complete years (every month represented at least 1 time in the year) I would use a window function in sql -- do you know a good way to query this in R? The closest thing I can think of is to use some sort of summarize len(year) and then make an assumption that I should have at least 300 datapoints/year, otherwise drop it
I’d do group_by, mutate and filter
Since you didn't want to hard code certain things in the title one could do this for the blue text as well:
value_today
Yes! I was thinking about this too 🤓
a smart move with the "glue" function and create the fake year '2022'
Friday being the 6th day is a Jewish plot! Ita called using the jewish:: package 😅