WHAT ARE LOGIT MODELS?? (Logistic models)
Vložit
- čas přidán 5. 09. 2024
- All my stats videos are found here: www.zstatistics...
See the whole regression series here:
• Regression series (10 ...
To download the jaybob.csv dataset, head over to the website above, I'll upload the data (and associated model worksheet) to the video page.
It takes a lot of effort and subject knowledge to keeps things simple. Thank you @zedstatistics for doing do :)
Hey! How do you convert percentage into binary?
i dont know what you are going through right now (since you are not active on this channel anymore) but i want to say this YOU ARE ONE OF A KIND in all youtube. its been 3 years since i got to know about this channel and i keep coming back everytime i need to brush up my concepts
superb! I wish you were my teacher back in school. i have nothing to do with statistics now but still I have come this far, to 4th part of your video, because of your explanation. They just kept me interested. Awesome! Please keep making such videos.
Believe me, not many students understand stats and not many teachers know how to teach stats. The way you teach is absolutely good and practically understandable. Thank you so much!
Hey! How do you convert percentage into binary output?
@@ammar46 step-1: first have the percentage ready in decimal format (e.g. 45.8%)
step-2: if you are using any programming language it will have a built-in function to convert decimal no. to binary no.
step-3: if you are writing a function yourself you need to understand how the left part of decimal is converted to binary and how the right part of decimal is converted to binary.
You can find this converstion math in any computer logic text book or video. Write your function accordingly.
Thank you Zed Statistics! I am really fascinated about how in depth you go about any topic and moreover make it simple for audience to understand, you are a real teacher...hats off!!!!
The best 18:24 minutes spent on stats this year. Amazing! 17:18 to understand OR 4.73 gasp!!!
Hey mate! Just dropping by to thank you. Thanks to you I passed this crazy test at school. You are a true hero!
Hi, at the time 11:14 , you mentioned, for a $1000 increase in price, the odds of selling a car decreases by 16%, while exp(-0.1730) is 0.84 which is 84%. Could you please explain
84% actually means a 16% reduction! If my weight this year is 84% of my weight last year, then I have achieved a 16% weight reduction (100%-84%). Hope that helps
Maybe a mathematically more intuitive explanation
Ln (p/1-p)= 0.396-0.173 (price)+1.555 (PS), then (p/1-p)= e^(0.396-0.173[price]+1.555[PS]) which can be rewrite p/1-p= e^0.396 x e^-173 (price) x e^1.555 (PS), hence the multiplicative effect of e^-173 or 0.84 on the odd of selling the car (p/1-p) with 1 extra unit of the price. Since this model is already presented in the context of post-hypothesis testing (remember the coefficient for the price in the model has a p value
@@zedstatistics Thanks! I had the same question, so glad you had replied to this and that I scrolled down comments to check :)
Thanks for sharing it helps me a lot in learning logistic regression for my coming exam
@15:05 where does 4.23 as denominator come from?
The line about it has (P+3.23P) on the left side. Factor out the "P" and you have P(1+3.23). So that's 4.23P on the left side. Then you're wanting to get P by itself, so divide by 4.23.
@@ssandok thank you Scott! I really appreciate you taking the time to explain that
I said that once in the past but I need to say it again ... your videos are just so valuable. I feel like I am getting university level information for free, it is actually to good to be true. I went to university so I have something to compare with, though it may be that my comparison is a particularly bad example ... I can't tell 🙂
One and only video which can understand very easily in the u tube about logit models ❤
just want to say a big thank you, thanks to your video I have aced all my courses and received a scholarship :) I'm so happy!
That's really great to hear! How have your studies been going?
i've recently found your channel, and I can't describe how well and using simple terms you describe some confusing concepts. thanks a lot!
SUPERB! i have never come across such amazing, informative and on point videos on stats! you are truly talented, thankyou for sharing! please keep the videos coming
Oh man, what were the ODDS that I'd come across your videos? 😎
But in all seriousness, your channel is amazing. Thanks for making and sharing everything!
You are a genius, you have managed to solve all my statistical headaches with this series. Thank you so much!!
Thanks zed!!! Ur explanation is far more clearer than my expensive uni 😂. Top notch 👌👌👌
You explain this so well I understand now, my teacher didn’t really explain it 😓 you saved my grade for my stats unit cheers mate
Explained with such simplicity and intuitiveness. Cheers to you zedstatistics!
Your videos are super helpful!
Can you do a video on dummy variables, effect coding, and ordinal logistic regression?
10:15
Odd Ratio = Multiplicative effect of one extra unit of X on the Odd of Y.
If price increase by 1k, odd of selling car will be 0.84x.
Your videos are superr helpful. Its been 6 years since you uploaded this video, but just wondering. Could you make a video on ordinal logistic regression? I have found very limited videos that explain it appropriately.
Thanks so much for making all these amazing videos! They are so helpful, and please continue making them! Looking forward to seeing your videos on logs!!
Hey! How do you convert percentage into binary output?
Dare I say, better than even StatQuest!
It was the best explanation of the Logistic regression.
Thank you
Amazing course. Highly recommended!!
thanks man, you gave me +A for my econometrics module
Absolutely excellent. Best explanation of this I've found.
Great stuff. do you have any videos explaining interactions (categorical and continuous independent variables) and their interpretation?
Helped me a lot, thank you so much for your great content! :)
Yoyoyoyoyoyoyoyoyo Mr Statistics in the HOOOOOUSE. Ty sir. We out here, regressing, estimating, LIVIN’. 🇲🇽🇲🇽🇲🇽
Great! The exercise at the end made it all clear. Thank you!
just helped me write my program evaluation thanks bro
Thank you so much for the explanations!
I have an exam project I need to hand in in 7 weeks time and I've learned a lot from your previous videos. Crossing my fingers that you'll have the fifth video out soon :o)
Greatings from Denmark
Hey I know it's been years, but I was curious. How did that exam project go?
insanely good video
Thank you @zedstatistics! I echo the sentiments of the others here: thank god for you otherwise I would likely be failing this part of my course! One thing though, I do not understand the part at 15:14 where you say - 'just some algebra...hope it's not too advanced!' Well, it is! Do you have a video that might explain this bit for me please?
Awesome videos, these have helped me understand regression so much better! I have a test coming up next week and I definitely feel a lot better about it. I'm praying that part 5 comes out before then lol
Hey I know it's been a long long time, but I saw your comment and was curious. How did that test go?
All your vids are extremely helpful. Thank you so much!
@zedstatistics At 17:31 you have shown how to did we come across the value 4.73 for pink slip. How do we derive the value 0.84 for price variable as in case of pink slip we calculated for scenarios with pink slip and without pink slip. How doe we do the same in case of price?
You just need to reverse the process shown above so keep the Pink Slip as a constant value of 1 (or 0 if you prefer) and adjust price to 1 [i.e. 0.396+(-0.173*1)+(1.55*1)] for the first calculation and 0 for the second i.e. [0.396+(-0.173*0)+(1.55*1)] (so just reversing what he showed when calculating 4.73)
Very nice video and well understood
Fantastic ! Thank you.
You are my favorite person rn
This was SOOOOOOO Helpful!
May god bless you. You are the best teacher. Keep up the good work buddy. :D
Gifted speaker. So clear.
You are a gem I must say.
What do you say when someone asks you to pause the video and give it a go? "Not today"
Amazing, thank you
Waiting so long for that, good job :)
Hey Justin, many thanks for the videos!
I just have one thing to confirm, among all regression assumptions, which one is applicable also for logistic regression? All of them? Or some of them are no longer required in case of logistic regression.
Great Video, thank u so much!!!
OMG. Thank you so much!!
Hi, can you make a video about mixed logit model!
Very useful and clear. Thank you very much!
Great video and super clear explanation!
wait but the intercept has a p value of 0.4 so how can we create an equation when the b0 is insignificant
Amazing explanation!
crystal clear. many thanks @
zedstatistics
Can I use logit regression if the error term doesn't follow logistic distribution?
Great video! When will the fifth video be out?
Out now @Lam Lam. Check it.
Great work.
Great Job. Very informative.
WOW! I just had a sudden realization!
One question:
If you applied log on the left side,shouldn't you apply it to the right side also?
Thanks!
Much love, FR!
This is so good, thank you so much!!
Great video thank you!
thank you so much! it was really helpful
thank you very much for this video
Tx sir
You're the best!
These videos are so good! Have you in some video discussed Poisson regression?
Certainly have! See my website here: www.zstatistics.com/videos#/distributions
Ah, sorry not poisson regression (yet!). But the poisson distribution :)
Can logit be used when the dependent variable is not binary, but there are more categories (eg as in the previous example with four category of car age)?
Also looking forward to "Regression V"
I searched 'logit-linear models' and found this. Is this the same thing as logit-linear models. ?
Hey! How do you convert percentage into binary?
You can do this in excel btw
Didnt know Jamie Lannister taught stats
I want to give u 10 likes
How if the categories are more than 2?
This would be called a multinomial logit model! Perhaps a video on this to come...
Thank you very much. If it possible, could you make the example of it using dependent variables that use likert scale? (Strongly agree, agree, etc) Thank you!
Link to MLE mentioned in the video: czcams.com/video/7kLHJ-F33GI/video.html
Can you please make video on Generalized Linear Models ? Thank you for these videos.
You explained in under 20 minutes what took my college lecturer 6 hours...AND WAY BETTER! Super disappointed with the quality of universities here.
The car is too expensive. $4,500? I can buy 3 iPhone 12.
Why the channel is named after the fascists?
Yo yo yo chill out dawg 🐶