Why Multicollinearity is Bad? What is Multicollinearity? How to detect and remove Multicollinearity
Вставка
- Опубліковано 31 лип 2024
- Multicollinearity is a phenomenon where two or more independent variables are highly correlated.
In other words, one predictor variable can be used to predict the value of another. This creates redundant information, skewing the results in a regression model.
============================
Do you want to learn from me?
Check my affordable mentorship program at : learnwith.campusx.in
============================
📱 Grow with us:
CampusX' LinkedIn: / campusx-official
CampusX on Instagram for daily tips: / campusx.official
My LinkedIn: / nitish-singh-03412789
Discord: / discord
👍If you find this video helpful, consider giving it a thumbs up and subscribing for more educational videos on data science!
💭Share your thoughts, experiences, or questions in the comments below. I love hearing from you!
⌚Time Stamps⌚
0:00 - 3:12 - What is Multicollinearity
3:13 - 10:30 - Why Multicollinearity is Bad?
10:31 - 12:14 - Types of Multicollinearity
12:15 - 16:15 - How to detect and remove
16:16 - 17:05 - Does it affect all ML algorithms?
Whenever a new video gets uploaded I leave everything and start watching. I'm addicted to this channel.
Thank you sir.
I don't know every time what to say! How can u explain the best & in simplest way possible every time ? How do u make awsome representation , simplest explaination with examples anyone can understand? How do u do that? Thanku so much for working hard. U really r very professional , knowledgeable , ur teaching skills r also the best. I m great fan of this channel. Hatts off....God bless u ! U r the best one so is The Campus X.
Couldn't agree more. Hands down Nitesh sir is best when it comes to explaining hard core topics. I was blown away when I saw the "Principal Component Analysis" lecture. Sir is true gem 💎 for us.
Thank you very much for this video, was stuck with this exact question and couldn't find the answer anywhere.
Really the video was amazing sir and all the concepts are cleared in depth.
Thank You Sir
Great video ....u cleared all the doubts Thank You
I love the way you deliver the content is awesome
Very well explained. Just what i needed to see in the best way possible. Thankyou !
Also, just a suggestion! if you could create a playlist adding all such imp interview questions videos together would be lovely and attract masses too.
Most helpful video I have come across on this topic. Thank you so much, sir
Sir I am your great fan the way you explain everything in your videos is awesome I tired reaching your team for the course but still I didn't had any update on it please help me with this sir thank you for such great videos your work is appreciated sir
Appreciate your efforts on making this video.. just to mention that you talked about parametric and non parametric models in the end. KNN comes under non parametric models.
yeah i also need clearity on this. can anyone help
very nice explanation sir aise hi vidio banate rho
Your videos are really helpful. I am facing this issue of unable to think of the next step while coding. If someone tells me what to do or i see the solution, it takes ne seconds to write the correct code. Any idea how I can overcome this barrier. Thanks again for the awesome uploads.
Can you please make a full playlist on time series forecasting arima,sarima etc
Nice sir.. Keep it up
Hi, Can you please let me know when you upload the other interview's video?
please upload other answers videos of machine learning interview questions
Is is ok to remove a column from a df which has high correlation with some other independent column in a classification problem? Or it should be done only in case of regression?
Sir my question is . distinction between partial and perfect multicollinearity can u explain me?
sir PCA and PCR is not same right? if right then what is pcr
Multicolinearity problem occurs only in regression problems or in classification problem as well ?
Hello sir. Sir I want to take your classes. How i can go for it ? Please reply me . And also i visited your website through LinkedIn but it didn't opened so please tell me . Thanks sir . You are awesome ❤️really.
Sir basically i want to do a project on ml environment but we don't know ml . So i want to do your classes.
What is VIF?
Hi sir, that was very helpful indeed. However, one question still remains..like, how it is impacting the algorithms??
It all depends on computation power and time. If more features are there with multicolinearity that mean many features are giving same information and it is better to skip those features. So less features mean computation time will be less which is good.
Is Polynomial regression(adding polynomial features)a structural multicollinearity?
stackoverflow.com/questions/67914111/doesnt-introduction-of-polynomial-features-lead-to-increased-collinearity#:~:text=Yes%20it%20adds%20multi%2Dcollinearity,3%20features%20into%20your%20model.
Hi Sir,
do we check Multicollinearity for classification problem as well ?
If your classification model is linear, ex: logistic regression, then you can check for multicolinearity
Unfortunately sports is also iq driven because most games require strategies but we can ignore
Aaj engineering ka last exam tha abh devotedly sari videos lgani hae aapke channel ki . Ek problem sir I'm facing every job under data science Askin for experience how to solve it
Do more interesting projects which aligns with the vacancy that you're applying for and add it in your resume. Don't refrain from applying even if the requirement says 2-3yrs of experience. Once they go through your resume there's a High chance that you'll be called up for an Interview and then on it's your show
@@abhimanyukspillai6572 got it with good projects i can apply to 2 3 years experience post as well thanks for advice
you are diamond
But sir why exactly multicolinearity affects the performance of regression algorithms(if only prediction matters to us).
I don't see any effect on the accuracy, although yes, it would take more processing power which could've been optimised but nothing else...
From what I understand, multicollinearity doesn't impact the predictive performance of regression models significantly. I believe that it makes it difficult to individually identify the contribution of each predictor variable to the response variable. So the impact is on interpretation than on prediction, to put it simply.