Boxplots are Awesome!!!
HTML-код
- Опубликовано: 21 авг 2024
- Boxplots are Awesome!!! Don't believe me? Check out the 'Quest.
For a complete index of all the StatQuest videos, check out:
statquest.org/...
If you'd like to support StatQuest, please consider...
Buying The StatQuest Illustrated Guide to Machine Learning!!!
PDF - statquest.gumr...
Paperback - www.amazon.com...
Kindle eBook - www.amazon.com...
Patreon: / statquest
...or...
RUclips Membership: / @statquest
...a cool StatQuest t-shirt or sweatshirt:
shop.spreadshi...
...buying one or two of my songs (or go large and get a whole album!)
joshuastarmer....
...or just donating to StatQuest!
www.paypal.me/...
Lastly, if you want to keep up with me as I research and create new StatQuests, follow me on twitter:
/ joshuastarmer
#statquest #boxplot #statistics
Support StatQuest by buying my book The StatQuest Illustrated Guide to Machine Learning or a Study Guide or Merch!!! statquest.org/statquest-store/
once upon a time I was trying to learn machine learning, Now I play Guitar.
bam! :)
totes bad to the bones!!!!!!!!!!!!!! :3
Hahahahaha
ADHD or something lol
I'm atm. watching your Statistic Fundamental Playlist and loved it so far, but this one I think was not really good. Would loved to see more about how the whiskers are defined, mild and extreme outliners. How the boxplot could give you ideas about distribution and maybe about how the IQR can be used to compare categorical data etc.
"Hello fellow children" hahaha. Thanks for explaining it so simply and in a funny way. Appreciate it.
Glad you liked it!
This is such a great explanation and sooooo hilarious in a nerdy, mathematician kind of way (that I truly appreciate). Well done!
Glad you enjoyed it!
It’s definitely got an Andy Kaufman style about it, which I love ☺️
This is really a great simple video to ensure students understand what they're analyzing.
Suggestion: It may be worthwhile to discuss how the whiskers are actually calculated to help students understand that outliers are values that are "potentially" outliers and that their interpretation depends on the situation. That way, maybe they would be less tempted to just say "oh they're outliers, I don't care about those".
In any case, thanks for another great video !
For example, in my course we calculate the step as h = 1.5 * (Q3 - Q1). But why ? What is the impact of calculating these whiskers in different ways ?
Perhaps explaining the interquartile range (IQR), without really focusing in the IQR itself, would fit in well after explaining that 50% of the data is in the box. Then, maybe it could help students focus on the concept of data being within certain ranges and how data outside those ranges data may be suspicious.
To explain the 1.5, it could help to mention the 68 95 99.7 rule and how it is an approximation of where data is situated. Similarly, 1.5 is an approximation of this so as to generally have 99% of data contained within the bounds.
Concepts I got from notably [here](medium.com/mytake/why-1-5-in-iqr-method-of-outlier-detection-5d07fdc82097) and [here](www.cs.uni.edu/~campbell/stat/normfact.html).
I see how your videos must become rabbit holes for you now 🤔
These are all excellent suggestions! Every video is a tradeoff between the main ideas and the details. This is why I often have one video that focuses on the "main ideas" and another video on the same topic that focuses on the details.
@@statquest Makes sense, thanks again for all you do :)
Finally, someone CLEARLY EXPLAINED why StatQuest is cool!
bam!
I appreciate your efforts to make videos. It takes a bunch of effort and sometimes goes unappreciated. Keep it up amigo.
Thank you very much! :)
so what are the whiskers for?
Data going from the minimum value all the way up to the first quartile. And data going from the 3rd quartile all the way up to the maximum value.
@@VaneyRio hey thanks man. Been so long forgot I asked.
Simple explanation that conveys a lot. Thank you
Glad you liked it!
Thanks Joshua! Very useful. I am very confused between probability and likelihood. A video explaining the difference using a simple example will be very useful.
ruclips.net/video/pYxNSUDSFH4/видео.html
wait so what do the whiskers represent?
He may have left it alone because it depends on who you ask. If there are no outliers, then it represents min + max of the data. If outlier(s) are present, then the values for Q1 - 1.5×IQR and Q3 + 1.5×IQR are the "fences" that mark off the "reasonable" values from the outlier values. Outliers lie outside the fences of the whiskers. (Josh, step in if you would like to define it more succinctly.) Great videos by the way.
UPDATE: Josh explains it in "StatQuest: Quantiles and Percentiles, Clearly Explained!!!"
ruclips.net/video/IFKQLDmRK0Y/видео.html
it captures the max and the min. Just pause at 1:32 and it all explains.
Hi, Josh, thank you for such an absolutely awesome video. Can you do a Boxplot clearly explained when you have time? Thank you again!
loved the thumbnail man XD
Thanks! :)
I like boxplots for when you look at "distribution" of CBCs, or weights, etc. Like that you can see the difference in the median, max and min for each condition...What do you think?
Sounds good to me! :)
Once upon a time I used to play guitar, now I play machine learning
bam!
Thank you a good vidéo to learn statistic science
Thank you! :)
I gave you a like just for the intro. No idea how the video is but you deserved it.
Thanks!
21 year old student looking at this to freshen it up lol
BAM! :)
student in mid 30s here ;) ! thanks Statquest, u rock!
hmmm carlin fan....BAM!!
:)
GREAT video!!!!!!!!! just a tiny question: many text books and teachers "ask" or more precisely "force" students to plot normal distributed data using bar plots. but i think boxplots present data bettter no matter the data is normal distributed or not. how do you see? and which do you choose?
Boxplots are so much better than barplots, especially when you can overlay them onto the raw data. They answer every question I have about the data.
@@statquest Thanks for answering!!!!!!!!!!! that helps a lot!!!!!!!!!!!!!!!!!
Hey! Thanks for the amazing video. May you explain as well the violin plot?
I'll keep that in mind!
can we give some credit for the singing tho
Please! :)
stupidity among people has a Normal distribution and in normal distribution Mean and Medain are the same, so G Carlin was technically right I think. BTW thank you for your awesome clear explanations.
noted
Bam !!
YES!
is there a longer version of this song hhh :D
Ha! I'd forgotten about this song. It's a good one. :)
Josh, could you explain - what is meant by A, B, C? If those are categories of measurements, why the dots are scattered round them?
A, B and C are just 3 different groups that we measured. For example, A = children, B = teenagers and C = adults. The dots are the raw data. For example, we might have measured how much time they spend on social media. The boxplots summarize these dots
@@statquest But why do dots belonging to group "A" not all lie on vertical line "A" ?
@@SergeySenigov So they can be seen. If they were all on a single line, they would overlap and it would be hard to tell what the original sample sizes were.
Josh, why don't you explain those fancy "violin plots" too? And their relationship with box plots and density plots. Please :)
I'll keep that topic in mind. :)
0:30 start
:)
Excellent....
Thank you! Cheers!
i still don't get what whiskers are supposed to represent...
They can represent different things, depending on how the boxplot is configured. Generally speaking, however, they give you a sense of how much more variation in there is in the data beyond the box.
there is an error, behind the median is 75%
What time point, minutes and seconds, are you referring to?
Hi can you make video playlist for probabilities concepts for machine learning.
Almost every video I have that covers probability and statistics applies to machine learning: statquest.org/video-index/
i love you.
Thanks!
i luv ur videos!!!!
Thank you! :)
"Totes bad to the bone!"
:)
1:18 median is a type of average, as well as mean and mode soooooooooooooo
;)
What is outlier here?
An outlier is a measurement that is very different from the rest and is often the result of some sort of mistake - like something was mislabeled.
How can boxplots be useful? What can we learn about a set of data from its boxplot?
I love box plots as a way to summarize large datasets. If you don't have much data, you can just plot the data and it's pretty easy to make sense of. But when there is a ton of data, you get lots of overlapping points and it's harder to make sense of it. That's where a boxplot comes in handy.
HAHAHA loved it
:)
I am taking Biostatistics and I hate it. Thanks for making this!
Good luck!
Awesome :D
no