About

Hi! I’m Josh Starmer and welcome to StatQuest! StatQuest started out as an attempt to explain statistics to my co-workers – who are all genetics researchers at UNC-Chapel Hill. They did these amazing experiments, but they didn’t always know what to do with the data they generated. That was my job. But I wanted them to understand that what I do isn’t magic – it’s actually quite simple. It only seems hard because it’s all wrapped up in confusing terminology and typically communicated using equations. I found that if I stripped away the terminology and communicated the concepts using pictures, it became easy to understand.

Over time I made more and more StatQuests and now it’s my passion on YouTube.

Advertisements

59 thoughts on “About

  1. hi,
    thanks for the great videos, I just discovered u! helpful and fun. BAM!!!

    I’m trying to look around here, and I feel a good index page of ALL posts (video + code for each) could be very helpful, or may be you have one and I just didn’t find it :)

    the “video index” page is missing a lot of stuff, that you can find only by scanning through the main home page. but the main home page is only organaize by date of posting, not by subject, and is very difficult to search.

    many videos that appear in main home page I can’t see in the video index page,
    and it is also difficult to fine the code sections of many of the videos

    your site is a Treasure Box! a good index page of ALL posts (video + code for each) would help us really make good use of it! (and if I just missed it and it exists… I sorry :) please guide me how to…

    thanks a lot!

    Liked by 2 people

  2. Just let you know you’ve taken online learning to the next level. I understand all the methods better than I did. I think now I can explain them to my grandmother just like Albert Einstein’s quote “You do not really understand something unless you can explain it to your grandmother.”

    Many thanks Josh

    Like

    • Thank you so much!!! I’m very happy to hear that my videos are so helpful. I have a lot of fun making them, but you telling me how much they have helped you learn about statistics – that makes it more than just fun. It’s meaningful.

      Like

  3. Your videos are really easy to understand . I usually start watching one when I eat lunch . You can understand how easy it is and how entertaining it is . It’s a lot of work making such video and great thanks to your efforts .

    Like

  4. Time for another big thank you and praise for your wonderful wit and teaching. I just went through the edgeR and DESeq videos and for dessert the t-SNE…great for the day after Thanksgiving…the series is terrific…

    Like

  5. Hi,

    Thank you for your videos, ability to explain material so clearly is a gift.

    I would add two companion videos (or pdf? mathjax in html page?) for each topic with

    a) math with derivations at undergraduate level (Lebesgue measure would probably be too deep, on the other hand, you can explain things really well, couple of videos about Lebesgue with lots of examples could be an interesting challenges and I believe, there will be a lot of people how would be grateful?)
    b) workout of the topic in python numpy/scipy or R

    I noticed that you do have some R and some math in the videos, but this is a bit too casual. Those additions would make your website self contained, that would be useful for autodidacts. But I do understand, that this would be a huge work and keep you probably busy for next couple of years, so may be it is too much to ask.
    Just an idea to consider

    once more, thank you very much, your videos are really good

    Like

    • Thanks so much for all of the ideas! I plan on doing StatQuest for as long as I can make videos – so, who knows? Maybe one day I’ll get around to this. I like the idea of people being able to go as deep as they want to go to learn about statistics and machine learning.

      Like

  6. Hi Josh, watched your youtube video on Benjamin Hochberg method and it’s pretty damn amazing (bam!) because it helped me to finally understand the algorithm. I really can’t wait to watch ALLLLL the remaining videos during my upcoming holidays. Keep rocking and keep doing what you’re doing, you make stats really sexy ;)

    Like

  7. I Was creating a capstone project for the final term. Your videos helped a lot. Your teaching method is really amazing. Thanks a lot.. From an Indian living in Canada.. BAM!!!

    Like

  8. Hello Josh,

    You are doing a great job BAM..
    The videos are very useful to understand the concept. The great thing about you is that you are sharing your knowledge without any cost i like that. If you need any help on creating the videos i can help you and contribute.

    Like

  9. Hey Josh!

    I would like to thank you in person for your Machine Learning videos. (BAM)
    Thanks to you, I got AA from Machine Learning course. (TRIPLE BAM)
    If I get a job, I promise I’ll buy your songs.

    Like

  10. Hi Josh,

    I know you don’t just walk up to Chez Panisse and ask for their souffle recipe, but can you, by chance, give us a peek into how you make your videos? That is, what techniques and software do you use to produce the mathematically accurate but visually pleasing diagrams, illustrations and animations? I’m a high school biology teacher hoping to do the same some day.

    Like

    • I’ll pull back the curtain and tell you that I started using PowerPoint and iMovie. I just drew everything by hand, except for a few graphs, here and there, drawn in R (a programming language for statistics). Then PowerPoint started crashing and causing a lot of other problems, so I moved to Keynote, a free Apple program (like iMovie). Keynote is way better than PowerPoint and it has a nice, built in equation editor, which is useful, but I avoid it as much as I can. If I make writing equations hard, then that forces me to use pictures to tell the story instead – and that’s what StatQuest is all about – using pictures to explain what is going on. The equations should only be there for very specific things.
      Anyway, long story short: there’s nothing special about the tools I use.

      Like

      • I just discovered your videos and simply love them. I have the same mission and job as you have, and although I have not made many videos, I have chosen an alternative medium for my statistical communication: Knitting projects. I am afraid my blog is still in Norwegian only, but you might have fun looking at the pictures :-) http://www.statistrikk.no

        Like

    • Awesome! I think I looked at that conference, or something like it, but was worried that it was geared towards people who teach statistics in a classroom. I don’t “officially” teach statistics. I just make videos as a fun thing to do in my spare time. When I’m at work I do biostatistics and computational stuff as a research collaborator. – so I was worried that I wouldn’t fit in with that group. But maybe one day I will. More likely, I’ll come to Norway to go hiking. I’ve heard it is amazing and always wanted to go.

      Like

      • -I just make videos as a fun thing to do in my spare time.
        How awesome isn’t that! Well, I will go to USCOTS this year to peel oranges :-) And if you plan a trip to Norway, you must let me know! Looking forward to more of your lovely videos!

        Like

  11. Hey Josh, do you think you could maybe make a video explaining covariance? I didn’t see any on your channel, and think it’d be very helpful!

    Like

  12. Hi, I recently found your videos on youtube and this website. I really like them. You deliver the materials in the way to be very easy to understand. I am now starting to study about single cell analysis. However, I am confused what materials I have to start with. Could you please give me lists of materials I should learn about single cell analysis? Thank you very much

    Like

  13. Hi Josh,

    First of all, I can’t thank you enough for your videos. It has made my life a lot easier. BA-BAM!!

    Secondly, do you think you could maybe make a video on time-series forecasting concepts like random-walk, serial correlation, stationarity, and ARIMA. I have been struggling to understand it for quite some time.

    Thanks!

    Like

  14. Hello Dr. Josh,

    I have become a fan of yours as well as statistics after visiting your channel. Is there any link to download slides you show in your video? A hard copy in hand would be very much helpful while watching your videos. Thank you.

    Like

  15. Had some questions related to the Adaboost example. After the first stump using Patient Weight (176), when we want to identify the second stump using the new sample data set (and weights 1/8), should we consider all the records with patient weight as 167 as 1 to find the average weights to be used in the Patient Weight stump?
    If I use only 1 instance of 167, the average weights come out to be (140.5, 161.5, 169.5 & 188.5). The Gini for Chest Pain and Patient Weight > 169.5 comes out to be equal. Which variable to use for the stump? Is the approach correct or am I missing something?

    Would appreciate a reply and let me know if more details are needed.

    Thanks

    Like

    • You use all 4 copies of the sample that has weight = 167 to make the new stump. This means there will be a large penalty if this sample is incorrectly classified since all 4 copies of it will be incorrectly classified. I talk about this at 18:21 in the video: https://youtu.be/LsK-xG1cLYA?t=1101

      Like

  16. Hi Josh,

    Just a quick question on your excellent PCA video (https://youtu.be/FgakZw6K1QQ?t=637) you show that the eigenvector for PC1 can be described by 4 parts Gene 1 and 1 part Gene 2. Does the direction matter? That is, can the eigenvector also be described as -4 parts Gene 1 and -1 part Gene 2?

    Ultimately, I’m guessing it doesn’t matter, as the data will still be separated on the final PCA plot relative to each other (just flipped?)

    Thanks for any advice, keep up the amazing videos.

    Jem

    Like

    • Direction does not matter. This means that the results can flip, just like you predict. I’ve seen this happen using different PCA functions applied to the same dataset.

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s