Wow, thank you for this video. Very concise with great explanation for practical use. I appreciate the clean and effective layout and your pacing. By far the most helpful video. The data I'm working with isn't inherently normally distributed and thus I work with it on the log scale for regression purposes. I was unclear as to whether I should take the average before logging or vice versa, but see that I didn't even consider how the geometric mean may be more relevant. In seeing if there was something similar for the geometric median, it seems that that is a whole other can of worms that I should probably ignore for the sake of methodological complexity hahah. Thank you for your work with these videos!
And thank you for the feedback! I really appreciate it! I'm very happy this was helpful for you. If there's anything else I can help with, just let me know
Why Arithemetic mean tends to be greater than goemetric mean while goemetric depends on compounded returns and arithmetic return depends on one time investment ? I cant understand that point can you declare it for me please.
The reply to this would take longer than I can fit in the comments, but as long as one of the numbers is positive and none of them is negative, the arithmetic mean is ALWAYS greater than the geometric mean. The Wikipedia page actually has a pretty good proof right at the top. Hope it helps. en.wikipedia.org/wiki/Inequality_of_arithmetic_and_geometric_means
Multiply all the numbers and take the nth root where n is how many numbers you multiply. 2 numbers will be a square root, 3 numbers a cube root, 4 numbers a 4th root and so on
Geometric mean is really used in an exponential system. In your case, the events could go either way. If you consider them independent, then arithmetic mean would be best. If you consider training to be a factor and want to figure something like the average percent improvement, geometric mean would be a good starting point. I have another video that talks about average rates of change that might be helpful
I ran into this at my job as a math tutor and didn't know what it meant. This was super helpful, thank you! :D
My pleasure!
Wow, thank you for this video. Very concise with great explanation for practical use. I appreciate the clean and effective layout and your pacing. By far the most helpful video. The data I'm working with isn't inherently normally distributed and thus I work with it on the log scale for regression purposes. I was unclear as to whether I should take the average before logging or vice versa, but see that I didn't even consider how the geometric mean may be more relevant. In seeing if there was something similar for the geometric median, it seems that that is a whole other can of worms that I should probably ignore for the sake of methodological complexity hahah. Thank you for your work with these videos!
And thank you for the feedback! I really appreciate it! I'm very happy this was helpful for you. If there's anything else I can help with, just let me know
You unlocked my mind
please make more videos
Adding more videos can help. More math tutor like. Me❣️❣️❣️
I'd be happy to! Do you have any particular topics of interest?
thank you so much! great explanation
Thank you!
Why Arithemetic mean tends to be greater than goemetric mean while goemetric depends on compounded returns and arithmetic return depends on one time investment ? I cant understand that point can you declare it for me please.
The reply to this would take longer than I can fit in the comments, but as long as one of the numbers is positive and none of them is negative, the arithmetic mean is ALWAYS greater than the geometric mean. The Wikipedia page actually has a pretty good proof right at the top. Hope it helps.
en.wikipedia.org/wiki/Inequality_of_arithmetic_and_geometric_means
Thanks. Very useful.
What if my data type contains 5 or 6 numbers?
Good question. If it contains 5 numbers, you'll multiply them and then take the 5th root. For 6, you'll multiply them and then take the 6th root
Thank you. That was well explained.
Explanation is very clear.. i understand
Thank you, I glad I could help!
Thanksyou somuch.. its a bighelp
Sometimes im very curious regarding square root🤣 sorry to tell about this can you add more and explanation regarding this matter
How can I find the Geometric mean?
Multiply all the numbers and take the nth root where n is how many numbers you multiply. 2 numbers will be a square root, 3 numbers a cube root, 4 numbers a 4th root and so on
Thanks
Thank u
Of course! I live to serve.
@@MrPledgerStaysAfterSchool love from india
What if you have a series of numbers that are increasing over time, hitting some limit (.e.g. race times of an athlete training to get faster)?
Geometric mean is really used in an exponential system. In your case, the events could go either way. If you consider them independent, then arithmetic mean would be best. If you consider training to be a factor and want to figure something like the average percent improvement, geometric mean would be a good starting point. I have another video that talks about average rates of change that might be helpful
cute