Thank you for this Innovation. I know many of us are lazy and tend to shy away from statistics but at least you made our life easy. I will share these videos with my students.
Very well explained. Thanks a lot. I would like to know if a video could be made on regression analysis with multiple independent variables and multiple dependent variables. Specifically when a dependent variables have more than 5 items.
Thanks for your question! I'll add it to the list, but may take time!!! We have this video, don't know if it helps yet ruclips.net/video/xVBwXqnWPyE/видео.html! With multiple dependent variables, it's actually just calculating multiple regression models with one dependent variable each again.
Excellent and fun explanation. I would like to test if a series of continuous IV predict two categorical variables (income level and gender). As you see, the DV are two, one of them is binary (gender) and the other has several levels (low, medium, high, extreme). What type of regression shall I use? I considered bivariate logistic regression, but here one of the DV is not binary.
Hello, thank you very much for your positive feedback! For the first case I would of course use a logistic regression, in the second case I would use a linear regression and see how the errors are distributed! Regards, Mathias
@@Jdonovanford As far as I know you always do multiple models when you have multiple dependent variables. Each dependent variable needs its own model, maybe you say at the end I have one model for the two dependent variables, but then there are actually two models.
If you want to do anything that tests for a probability, then logistics regression is good. For example, probability something is spam or not. Probability a transaction is declined or not based on your purchase habits. Probability a patient has a disease based on a list of symptoms. These are some examples of how you could use the model. Almost anything, with any number of variables can be used in the model.
Hi dear! Would like to ask you a question. My DV is categorical- It is representative of whether participants committed = 2 harsh accelerations during a drive. My first IV is age (continuous) My second IV is total sleep time in the 24 hours prior to the drive (continuous) Example of how my data looks: Participant ID/ Harsh accelerations/ Age(years) /Total time (mins) 1 2 42 500 1 5 42 540 2 6 23 620 2 2 23 450 2 2 23 420 3 1 31 200 3 11 31 280 3 4 31 340 3 1 31 360 In using a binary logistic regression (on SPSS) to examine whether driver age and total sleep time (TST) in the 24 hours prior to shifts had an impact on whether drivers would have
![Logistic Regression [Simply explained]](http://i.ytimg.com/vi/C5268D9t9Ak/mqdefault.jpg)
![Logistic Regression [Simply explained]](/img/tr.png)
![Causality [Simply explained]](http://i.ytimg.com/vi/bm6V84Lgz2w/mqdefault.jpg)
After 10 hours trying to understand this concept, I'm finally found a solution here. Thank you a lot 👏
Thanks for the Feedback: ) regards mathias
OMG THANK GOD FOR THIS VIDEO YOU MADE IT SO EASY TO UNDERSTAND FOR MY STATS EXAM
Thanks!
Thank you for this Innovation. I know many of us are lazy and tend to shy away from statistics but at least you made our life easy. I will share these videos with my students.
Many many thanks!!! Regards, hannah
@@datatab Thanks you Hannah I look forward to being your student number one👋
Easy to follow and very helpful!
Glad it was helpful!
Superb. Love the woman's voice, too.
Thanks : )
Very helpful!
The best youtube video I played in 2022.
Glad it was helpful and many thanks for your nice feedback! Regards, Hannah
Very well explained. Thanks a lot. I would like to know if a video could be made on regression analysis with multiple independent variables and multiple dependent variables. Specifically when a dependent variables have more than 5 items.
Thanks for your question! I'll add it to the list, but may take time!!! We have this video, don't know if it helps yet ruclips.net/video/xVBwXqnWPyE/видео.html! With multiple dependent variables, it's actually just calculating multiple regression models with one dependent variable each again.
Thank you so much for simplifying this, now it looks like drinking water
You are the best Hannah :-) :-)
Many thanks : )
Very well explained.....thanks
Thanks for your Feedback!
Outstanding lecture , weldon
Thanks for your Feedback!
Nice work carry on plz.
Thanks!
After entering the independent and dependent variable ,the answers are hidden.How can I use this logistic regression calculator?Should I download it ?
Hi, sorry, unfortunately we can't offer DATAtab for free, you just have to buy it and then it goes online!
Excellent and fun explanation. I would like to test if a series of continuous IV predict two categorical variables (income level and gender). As you see, the DV are two, one of them is binary (gender) and the other has several levels (low, medium, high, extreme). What type of regression shall I use? I considered bivariate logistic regression, but here one of the DV is not binary.
Hello, thank you very much for your positive feedback! For the first case I would of course use a logistic regression, in the second case I would use a linear regression and see how the errors are distributed! Regards, Mathias
@@datatab Thanks, I just wonder if one model can handle the two variables at the same time. You seem to suggest doing two separate analyses.
@@Jdonovanford As far as I know you always do multiple models when you have multiple dependent variables. Each dependent variable needs its own model, maybe you say at the end I have one model for the two dependent variables, but then there are actually two models.
@@datatab interesting. Pardon my ignorance. So multivariate models are usually basically univariate ones in mass?
In any case, for the second case I wouldn't use linear regression becasue the DV is categorical. How about multinomial logistic?
Ma'am can logistic regression use to find the impact of risk management strategies on company performance?
Yes, if your data meets the requirements you can do it. 😊
Thank u soo much great work keep it up
@@Thath97 thanks for the nice Feedback!
If you want to do anything that tests for a probability, then logistics regression is good.
For example, probability something is spam or not.
Probability a transaction is declined or not based on your purchase habits.
Probability a patient has a disease based on a list of symptoms.
These are some examples of how you could use the model. Almost anything, with any number of variables can be used in the model.
you can play at x1.5
: )
Thank you for your video it’s amazing but can I use logistic regression of small sample size n=12?
Unfortunately, I can't give you a general answer, but 12 is very few! Regards Hannah and Mathias
@@datatab Thank you for answering. I hope anyone know the appropriate way of make a regression for non- parametric analysis inform me
Hi dear! Would like to ask you a question.
My DV is categorical- It is representative of whether participants committed = 2 harsh accelerations during a drive.
My first IV is age (continuous)
My second IV is total sleep time in the 24 hours prior to the drive (continuous)
Example of how my data looks:
Participant ID/ Harsh accelerations/ Age(years) /Total time (mins)
1 2 42 500
1 5 42 540
2 6 23 620
2 2 23 450
2 2 23 420
3 1 31 200
3 11 31 280
3 4 31 340
3 1 31 360
In using a binary logistic regression (on SPSS) to examine whether driver age and total sleep time (TST) in the 24 hours prior to shifts had an impact on whether drivers would have
Hello, unfortunately I am completely full right now and therefore can not answer any questions. Especially not about SPSS, sorry
2 negri chicken woni chicken woni chicken 2 negri
???
@@datatab they're probably just being racist against the lecturers accent. Just delete the comment.