The samples actually will be bimodal, but the sample AVERAGES will be approximately normally distributed because the average is always in the middle of the two peaks as a measure of center. In fact, the reason the sample averages vary at all is purely because of random deviation, not because of the bimodal population. And random variation always has a normal bell curve shape.
For example, average of binomial distribution will be any value in the interval between 0 and 1: e.g. 0,32; 0,47; 0,78 .... Then, these averages are not distributed as taking only two possible values, and they would follow normal distribution.
u know ,binomial distribution is the sum of n independent bernoulli distribution with the same probability of success. there is another version of central limit theorem, that is the sum of n independent variables with same expectation and variance will approximate to normal distribution when n is large. so , when n is large and p is relative small , the distribution of binomial can be approximated by normal distribution by following the central limit theorem.
@@darkhoof69 so say the dragon wingspan on one peak is 20 metres and on the other it is 42 when we calculate the mean it is 62/2 which is 31, thus the peak in the centre of the curve is 31 and is shown to be one individual shaped curve, is that right?
OMG!!!! This is well-explained 😀I'll give an A+ rating for this ....... If only dynamics could be explained like this !!!😔....anyway ...thank you for this video❤
If the underlying distribution is normal, the sampling distribution is normal for any sample size and doesn't become more normal with increasing sample size as claimed. This only holds if the underlying distribution is not normal. But then, in fact, no distribution in reality is perfectly normal (certainly rabbit sizes are not, as they cannot be negative). Note also that in practice the Central Limit Theorem may fail because of violations of the "identical and independent" assumption and for data quality reasons (for example with outliers).
Not quite! The t-distribution is for sample means (minus the population mean) DIVIDED BY sample standard deviations. The t-distribution arises because the sample SD is a noisy estimate of the population SD. So even if [xbar - mu] has a Normal distribution, the ratio [xbar - mu]/s has heavier tails than a Normal (more likely to get large positive or large negative values), and the t-distribution accounts for these heavier tails.
ok, but you left unexplained the most interesting feature of it! this video gives no insight on why &/or how we can use normal distribution to measure variables which are not normally distributed! if this is trivial, your entire argument is just tautological, and perhaps this is why you covered up its very core...
pov: your statistics teacher sent u here
scary accurate
LMFAO
tru
statistics teacher planning to send students here
My statistic teacher played this in class.
I need all my statistics explained like this.
It makes Statistics interesting. It really helps me to teach my students about the concept of normal distribution. Thank you very much!
This is fun, cute, and a great explanation!
Thank you! The central limit theorem explained clearly, and cutely. :)
Thank you. This is the ONLY youtube explanatory video I could understand, lol.
Clear explanation & entertaining. Like it!
Great ! helped me a lot in my biostat class
I don't get why the bimodal distribution still turns into a normal distribution for samples... Anyone?
The samples actually will be bimodal, but the sample AVERAGES will be approximately normally distributed because the average is always in the middle of the two peaks as a measure of center. In fact, the reason the sample averages vary at all is purely because of random deviation, not because of the bimodal population. And random variation always has a normal bell curve shape.
darkhoof69 Is a condition of the universe, same as speed of light and Fibonacci sequence.
For example, average of binomial distribution will be any value in the interval between 0 and 1: e.g. 0,32; 0,47; 0,78 .... Then, these averages are not distributed as taking only two possible values, and they would follow normal distribution.
u know ,binomial distribution is the sum of n independent bernoulli distribution with the same probability of success. there is another version of central limit theorem, that is the sum of n independent variables with same expectation and variance will approximate to normal distribution when n is large. so , when n is large and p is relative small , the distribution of binomial can be approximated by normal distribution by following the central limit theorem.
@@darkhoof69 so say the dragon wingspan on one peak is 20 metres and on the other it is 42 when we calculate the mean it is 62/2 which is 31, thus the peak in the centre of the curve is 31 and is shown to be one individual shaped curve, is that right?
OMG!!!! This is well-explained 😀I'll give an A+ rating for this ....... If only dynamics could be explained like this !!!😔....anyway ...thank you for this video❤
This was so wonderful!
Thank you so much for such a simpler explanation
If the underlying distribution is normal, the sampling distribution is normal for any sample size and doesn't become more normal with increasing sample size as claimed. This only holds if the underlying distribution is not normal. But then, in fact, no distribution in reality is perfectly normal (certainly rabbit sizes are not, as they cannot be negative). Note also that in practice the Central Limit Theorem may fail because of violations of the "identical and independent" assumption and for data quality reasons (for example with outliers).
This has to be the future of learning.
This was super cute and easy to understand! Thank you!
Loved this!
I was searching for bunny dragon hybrid babies but this is really cool too!
Amazing 🔥
my statistics prof made me watch this
Who else is here because they're studying Psychology at University?
I'm studying political science & international relations, our asst. professor played this to us in class.
I LOVE THOSE PİKA-BUNNİES.
@@mansur_ali I'm here for stats
*Stats my BOIISSS*
Audiology student here🙋
Great video, very helpful.
I want more!!!!
MAT183?
So cute!!!!
great video - the only problem is: means, at least for smaller samples, have Student's t distribution an not NV.
Not quite! The t-distribution is for sample means (minus the population mean) DIVIDED BY sample standard deviations. The t-distribution arises because the sample SD is a noisy estimate of the population SD. So even if [xbar - mu] has a Normal distribution, the ratio [xbar - mu]/s has heavier tails than a Normal (more likely to get large positive or large negative values), and the t-distribution accounts for these heavier tails.
This is great
Love it!
love it!
Who else is here because of Stats in University of Pretoria
STK110 man
very helpful for the likes of me, TY!
Is there a form to know when the sample's mean comes from a nonnormal population?
Bachtel sent me
Hi joe
interesting
when did thee new york times start reporting on dragons as statistics?
ok, but you left unexplained the most interesting feature of it! this video gives no insight on why &/or how we can use normal distribution to measure variables which are not normally distributed! if this is trivial, your entire argument is just tautological, and perhaps this is why you covered up its very core...
çağatay edemen brought me here
any hkust homies??
Wer da wegem Luchsi? 😳😳😳😳😳😳😳
ISOM 2500 send me here LOL
Hey what's up STATS 250
*I see students are joining...*
RICARDO MAYER ME TRAJISTE ACA!!!! AGUANTE UDP
EIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
Hall 4th period wya?
MVHS?
they look like pikachu
Cy
who here from marianopolis lmfao
I wish I could get a refund of my fees and give it to you