Statistics 101: Model Building Methods - Forward, Backward, Stepwise, and Subsets
Вставка
- Опубліковано 22 лип 2024
- In this Statistics 101 video, we look at an overview of four common techniques used when building basic regression models: Forward, Backward, Stepwise, and Best Subsets Regression. The video is purely conceptual, no math, no software, or anything numerical. We will get to that. For now, I just want you to get a solid understanding of the big picture before we get into the weeds and specifics of each method. Enjoy!
My playlist table of contents, Video Companion Guide PDF documents, and file downloads can be found on my website: www.bcfoltz.com
For more info on XLSTAT: www.xlstat.com/en/
Happy learning!
#statistics #machinelearning #datascience
Awesome thanks, i'm still at the beginning of playlist 4, but this new video, gave me an incredible insight of what is coming, and that, changes everything. Thank you so much.
You are so good, truly doing humanity a great service. Thank you so much Brandon.
Thank you very much (once again) for your amazaing videos, @Brandon!
This video is a real thing for understanding some of ML concepts! Looking forward to seeing the next videos :). Congratulations!
Thank you for the video, great way of explaining complex models into simple ideas
Thank you for this. This definitely helped me understand this topic better especially with the visualization.
You have a great way of simplifying complex ideas and helping connect our understanding of the best uses for each type of model. Understanding this topic first helps me solidify the concepts. Thank you!
You're very welcome!
Dear Sir, I am delighted with your explanations. Thank you for everything you do for us.
And I must agree, excellent teacher! Sure hope you see my comment below, I love serendipity, and we could be on the precipice of it...This will probably be my last post about the subject, I wouldn't want to appear obnoxious with multiple entries regarding the same question. Thanks again for making Statistics something I can get my head around. It's certainly a gift, thanks for sharing it!
Beautifully explained!
Amazing teacher. Thank you so much!
This is a brilliant explanation of model building methods.
Thank you very much sir !!
Please keep sharing knowledge!!
Loving the way you approach to these topics. Definitely, the key factor that helps to understand these concepts and ideas is the visualization. I mean, I am studying for my final year Statistics by referring through 'Introduction to Linear Regression' by Montgomery, but what I always found hard was to relate it in reality, to visualize it, and your videos now make it much simpler to understand the same things as I have a picture or something in my mind and can relate to what the author is trying to explain.
Could you please come up with examples for each method I.e. Forward selection , backward deletion and stepwise method. Also with the AIC and BIC Criteria. Also would like to get more undersanding about Studentized Residuals and Standardized Residuals
Best stats videos I have come across! Really excited for the next videos in the series. Just one thing: can you add sources?
Keep it up! the way you explain is fantastic!
Great Video! Great Content, thanks for making it!
Very amazing video. Thanks for this contribution to learning
Awesome! many thanks 🤓
Great video actually. The visual explainer provides a lot of intuition before the maths is introduced.
You've explained excellent.thanks
this video is so useful and easy to understand
This is so good. Thanks for the video :)
great teacher!
Thank you! 😃
Hello Brandon, I want to thank you for all your videos. They were very useful for understanding better statistics. I was wondering if you did one video of the Linear Mixed Model. I'm really interested in learning that.
You area great teacher. I am going to watch all of your videos
Thanks. Very helpful video.
very informative
thanks
Thank you so much.
These are hands down the best stats videos on the Internet
Very kind of you! Thank you! I am glad the channel has been of some help to you along your learning journey.
Wow what a teacher!!
Always great
Been trying to learn this for weeks, but it clicked immediately when i watched this video
I enjoyed this. Was very easy to understand
Glad you enjoyed it!
Very good vid!
Very helpful. thanks
Hello. Thanks so much for your time and dedication. These videos are lifesaving. Are you planning to create companion slides for the ML videos? Thanks so much.
I found you because of my bewilderment of Statistics, and you have been most helpful. I wrote this on a comment from a video from 3 years ago, so if you see this twice, well, UA-cam bewilders me too sometimes! Mr Foltz, by any chance are you related to a Joseph Foltz? In 1964 he would have been a marketing executive at Marathon Oil and living in Lima. Sure would be a coincidence if you are, trust me. Of the the Statistics 101 videos out there, and I picked yours? If not, at least I found a terrific resource, and I'm grateful
Great video!
Glad you enjoyed it
very good my teacher
Love this channel ❤️❤️❤️❤️❤️❤️
You are so kind. Thank you. :) Glad you find it helpful.
Amazing
thanks
thanks very much, your youtube channel is such a blessing! please can we have a series statistical method on the repeated measure?
Thanks Joyce! I do have a video on repeated measure ANOVA. Is that what you are referring to?
Anyone else noticed that the thumbs up button lights up at 13:50 when he says "give it a thumbs up"? :)
Namaste !!
You just saved me from failing ♥️
Can you make a video on Bayesian models
Is this the one used in wrapper based feature selection in machine learning?
in 16:19 what number do you look at to consider if adding X3 is significant change
🙏 Namaste Ji
Namasteyyyy!!😄
Namaste 🙏
Namaste 🙏😍
can you reply in your comments from your old videos?
Namaste
I know you mentioned that different models and techniques comes with controversies, however I think you should at least mention what those are! These methods are generally frowned upon when doing statistical inference, as it can be seen as a form of p-hacking. For pure prediction and model building I'd say it's almost a different ballgame, and in that case you can use things like this.
I like that....grab a drink
why idid he say namaste?
Nice video. You began it with two questions: which model is best and simplest? how do we decide 'best' model?
At the end, you did not address or even mention those questions. This makes your video open ended and incomplete.