Hi Justin, I posted a lengthy question here yesterday regarding mortality rates, but I was able to work through to the answer. I’m leaving this follow-up comment though to thank you for all of your efforts with these tutorials. 😊
My favorite stats teacher by far! Thank you. One nitpick was your exerciser vs couch potato RCT example. The confusing part about that is that even if you randomize and make exercisers become inactive, it will be tough to drowned out the noise of the long lasting impacts of that lifestyle. If the sample was massive, I guess it would all average out and you could see some things. But I'd still expect there to be a large coefficient of variation...
Thanks Yusuf! Ha yeah by "couch potato" I was simply meaning those that were directed not to exercise. Could have been a little clearer :) Thanks for the feedback!
Really good examples of confounding. This is an important topic right now with the public suddenly so interested in the slew of correlational research on COVID, and reporters chasing miracle cures indicated by correlations.
Haven't seen a new Jeremy's Iron podcast out recently :( Would I be correct to assume that this isolation period has made it significantly more difficult to get together and record them? Love your vids!
Hi Zed, thanks for your video. I learned a lot from your regression videos and I have a question regarding the confounding variable and Exogeneity. Since a confounding variable affects both an Independent variable X and a dependent variable Y. Is there any chance it becomes an omitted variable and its explaining ability has been added to the error term in our regression model to cause the endogeneity problem? What if we add this confounding variable into our model to fix the endogeneity? Will it cause collinearity? Because it seems the confounding variable is highly correlated with one independent variable.
Wow! The simplicity of your explanation is out of the world!
Hello Justin. Amazing videos! Just started public health and it has been helping me a lot in my biostatistics module! Thank you so much
That was a great explanation of confounding. You have a gift. I’m enjoying all your videos.
Thank you! Your teaching style is easy to follow.
Hi Justin, I posted a lengthy question here yesterday regarding mortality rates, but I was able to work through to the answer. I’m leaving this follow-up comment though to thank you for all of your efforts with these tutorials. 😊
Thanks, Justin! Awesome explanation!
Please make more medical statistics. Very very important
My favorite stats teacher by far! Thank you. One nitpick was your exerciser vs couch potato RCT example. The confusing part about that is that even if you randomize and make exercisers become inactive, it will be tough to drowned out the noise of the long lasting impacts of that lifestyle. If the sample was massive, I guess it would all average out and you could see some things. But I'd still expect there to be a large coefficient of variation...
Thanks Yusuf! Ha yeah by "couch potato" I was simply meaning those that were directed not to exercise. Could have been a little clearer :) Thanks for the feedback!
Very nice presentation Sir. 👍
Thank you for the video. You're helping me a lot
You are amazing!!! I wish I watched yours first. Thanks, I love your accent too😊
perfect teaching
Well explained!
Really good examples of confounding. This is an important topic right now with the public suddenly so interested in the slew of correlational research on COVID, and reporters chasing miracle cures indicated by correlations.
True that, SD!
thank you
saved me right before my quiz tyyyy :)
Haven't seen a new Jeremy's Iron podcast out recently :( Would I be correct to assume that this isolation period has made it significantly more difficult to get together and record them? Love your vids!
Stay tuned Sally! New epsiode coming soon... and there's a podcast name change coming too. Something a little less cryptic :)
@@zedstatistics Super stoked to hear that!! Can't wait for it!!
Hi Zed, thanks for your video. I learned a lot from your regression videos and I have a question regarding the confounding variable and Exogeneity. Since a confounding variable affects both an Independent variable X and a dependent variable Y. Is there any chance it becomes an omitted variable and its explaining ability has been added to the error term in our regression model to cause the endogeneity problem? What if we add this confounding variable into our model to fix the endogeneity? Will it cause collinearity? Because it seems the confounding variable is highly correlated with one independent variable.
So what you basically do being a statistician ? You data scientist or just a nerd of stats?
Bro can you please suggest me some statistics books in which a large number of solved examples are present and i am a beginner
Thank you.
Are confounding and independent variables the same?
They can be. But in addition to just being another IV, they're an IV that affects both the outcome variable AND other IVs.
When there will be a movie about mediators and moderators?:)
Somehow we still need math language to reiterate everything. Please consider making separate videos including all the math.
Thank you.