My friend, I want to thank you a lot! for posting this videos. You are an excellent professor and communicator! My statistics profesor is the worse... I can't understand anything at all in class.. I don't even know why it is so expensive to pay for her class 4,500 a semester when all she does is talking and drawing numbers and we are all like 0_o ... Anyway, I really just wanted to thank you a lot for your help. This is my last semester as an undergrad and without these videos I don't think I could pass this subject and graduate. Thank you Thank you Thank you!
your videos are really useful, easy to understand rather than reading my lecture notes. especially the sampling distribution series, it helps me a lot. moreover, you have a clear voice, unlike the others. THANK YOU VERY MUCH!!!!!
These are by far the best statistics videos on youtube! Thanks to you i now feel that i should be able to pass my stats exam. Thank you so much! Please keep up the good work!
@@jbstatistics Having studied Stats at uni and having passed the exam but remained confused Well done! The very opposite of going from the notes of the lecturer to the notes of the student without going through the brain of either.
That which is particukarily noticeable is the time you take to really explain things in a slow relaxed way Well Done. Finally ubderstand the student t test.
Thanks, I'm glad you like my videos. Thanks for the suggestions. Gamma, Weibull, and lognormal are definitely somewhere on the horizon, but it will be a little while before I can get to them. (I've got a number of other videos lined up right now.) Cheers.
Thank you thank you thank you. My online university provides a link to your videos and I have been using them since the beginning of my stats course. I would be lost without them!
I really like your way in explaining its very simple and understandable. I hope that you create a video tutorial for Gamma, Weibull and Lognormal distributions because I didn't find any good video tutorials for these concepts. Thanks in advance =)
The terminology really fucks me over. I don't know when to implement certain formulas so that's a real hinderance although, this has sort of widened my perspective so cheers mate
My teacher never teaches in class so i'm forced to scour the internet and search for problems similar to the work he gives out to help me understand it better. After watching this video, I still am at a lost for what to do
Jenny FromDaBloc It's an area under the standard normal curve, and it can be found with software or a standard normal table. I have videos that show how to use the standard normal table. Cheers.
Take the Z value (at 0.842) and subtract it from 0.5. Because we want the probability above 23grams (at least) and the Z value 0.842 represents the area from 0 to 0.842 you need to subtract it from the whole area (which is 0.5 on the normal distribution table).
@8:39 why do we choose the middle data? We have to draw a sample of size 4 so we can take four values from the left or right as well? And will the mean will be the same as 21.4 for all samples we draw from this distribution?
I don't know what you mean by "why do we choose the middle data". We're randomly picking 4 values from that distribution; they'll be whatever they'll be. It's 4 random draws from that distribution. The distribution in green is the distribution of the mean of 4 randomly picked values from the distribution in white. Yes, each of the 4 has exactly the same distribution (the distribution in white).
@@letseconomics2938 You're not understanding what's happening. We're not choosing any specific values, or letting our judgement decide which side we want to grab them from. We're randomly picking 4 values from the distribution in white. If we do that, then the distribution of the mean of those 4 randomly picked values is the green distribution. That's just how the math works out. The green distribution has a lower variance for the reasons I discuss in the earlier part of the video.
I'll just add that sure, we might end up randomly picking 4 extreme values from one tail of the distribution from which we're sampling, that will have non-zero probability of occurring. But the probability of that is very small and is reflected in the sampling distribution of the sample mean.
I really like your way of explaining concepts. But if in a real case scenario where we don't have mu value(suppose a very large sample) how do we even calculate the probability??
Thanks for the compliment. As you know, we don't (generally) know mu in real life. So then we can't calculate a probability this way. The primary purpose of the discussion in this video is that we later turn this problem on its head, and use a known value of X bar to say something about the unknown value of mu. Working with the mathematical concepts discussed in this video (and some others), we can come up with an appropriate formula for a confidence interval for mu, and an appropriate test statistic to use in hypothesis testing. Statistical inference is built on concepts related to the sampling distribution of the estimator of a parameter. Edit: Which is what I allude to at the end of the video.
I have one doubt , x' mean is equal to population mean when we take mean of the sampling means otherwise not ? but your told that only one sample mean is equal to the population mean how tell me ?
P(Z>1.648) can be found by looking up Z of 1.648 on standard normal table. Area under the standard normal curve to the RIGHT of Z is ~0.954. Area to the LEFT of Z is what we are looking for, so P(Z>1.648)= 1(total area under std. norm. curve) - 0.954 = 0.046
I talk about this in detail in my other videos. When sampling from a normally distributed population, the random variable Z = (X bar - mu)/(sigma/sqrt(n)) has the standard normal distribution. When sigma is replaced with the sample standard deviation S, the quantity T = (X bar - mu)/(S/sqrt(n)) has a t distribution with n-1 degrees of freedom. This is always true when we are sampling from a normally distributed population, regardless of the sample size.
I am not sure what you are asking me. The standard deviation of the sampling distribution of the sample mean is just the square root of the variance of the sampling distribution of the sample mean. The standard deviation of the sample mean and the variance of the sample mean both decrease as the sample size increases.
That's ok I realised I read my lecture slides wrong. Was confused as to why they were saying the SD and variance increased with sample size, however they were saying the opposite. Thanks for actually replying though :)
Can somebody please explain how to properly estimate the population parameters: true mu and true sigma, when we don't have them? Can I use the sample mean as a good point estimator for the population mean? Can I use the sample standard deviation as a good point estimator for the true sigma?
In short, no. We might conceivably have an estimate of sigma, which could give us a rough estimate of the probabilities, but without knowledge of the value of sigma the exact probabilities can't be calculated.
The area to the right of 0.842 under the standard normal curve is 0.200. That's found with any statistical software, such as R, or by using a standard normal table.
Of what part? The mean and variance? The fact that if we're sampling from a normally distributed population, then the sample mean is normally distributed? Of the central limit theorem? I derive the mean and variance of the sampling distribution here: ua-cam.com/video/7mYDHbrLEQo/v-deo.html. I don't have video proofs of the other parts yet.
thank god for these videos. I learned in 12 min what I didnt understand in 2 days worth of class
My friend, I want to thank you a lot! for posting this videos. You are an excellent professor and communicator!
My statistics profesor is the worse... I can't understand anything at all in class.. I don't even know why it is so expensive to pay for her class 4,500 a semester when all she does is talking and drawing numbers and we are all like 0_o ...
Anyway, I really just wanted to thank you a lot for your help. This is my last semester as an undergrad and without these videos I don't think I could pass this subject and graduate.
Thank you Thank you Thank you!
+dasgomezkanal You are very welcome!
best stats tutor on the internet . Well done !
Thanks!
CAN I JUST SAY THAT YOUR TUTORIALS ARE THE BEST AND EASIEST TO UNDERSTAND OUT THERE
Thanks!
your videos are really useful, easy to understand rather than reading my lecture notes.
especially the sampling distribution series, it helps me a lot. moreover, you have a clear voice, unlike the others. THANK YOU VERY MUCH!!!!!
+Hugo Chong You are very welcome Hugo!
These are by far the best statistics videos on youtube! Thanks to you i now feel that i should be able to pass my stats exam. Thank you so much! Please keep up the good work!
Thanks Arian. I'm very glad you've found my videos helpful. Best of luck on your stats exam!
@@jbstatistics Having studied Stats at uni and having passed the exam but remained confused Well done! The very opposite of going from the notes of the lecturer to the notes of the student without going through the brain of either.
That which is particukarily noticeable is the time you take to really explain things in a slow relaxed way Well Done. Finally ubderstand the student t test.
Thanks, I'm glad you like my videos.
Thanks for the suggestions. Gamma, Weibull, and lognormal are definitely somewhere on the horizon, but it will be a little while before I can get to them. (I've got a number of other videos lined up right now.) Cheers.
Thank you thank you thank you. My online university provides a link to your videos and I have been using them since the beginning of my stats course. I would be lost without them!
+Cat Garcia You are very welcome! I'm glad you've found my videos helpful!
Those are simply the best explanations!
I'll watch all your videos.
+alexandra-stefania moloiu Thanks!
I love your videos! They're so informative and direct to the point! Plus, your voice is very clear and understandable! Thank you! +1 subscribe! :)
Thanks for the compliments!
excellent, extremely good
I really like your way in explaining its very simple and understandable.
I hope that you create a video tutorial for Gamma, Weibull and Lognormal distributions because I didn't find any good video tutorials for these concepts.
Thanks in advance =)
Very informative , crisp and clear.. Watched other videos but understood very well from your videos the concepts well. Thank you.
Thanks for the compliments!
Awesome video. This provided great help for my upcoming stats exam!
this is the video that made everything make sense in my head
the GOAT of stats tutor online
Thanks!
Best stats videos on youtube!
***** Thanks!
Your videos are a true life saver
The terminology really fucks me over. I don't know when to implement certain formulas so that's a real hinderance although, this has sort of widened my perspective so cheers mate
Love these videos!!!!! You all are awesome
Thanks! I'm glad to hear that!
My teacher never teaches in class so i'm forced to scour the internet and search for problems similar to the work he gives out to help me understand it better. After watching this video, I still am at a lost for what to do
Thank you, apparently you are a better instructor than my online applied statistics instructor...
You are very welcome.
Very clear explanation.
Thanks!
You save lives my friend
It helps me a lot. Thank you so much po.
That was *GREAT.* Thank you!!
Incredibly well done, thank you
Really useful! Thanks a lot!
A great help! God bless you!
Thanks!
Thank you JB! Excellent tutorials :-)
Thanks a lot. It is helpful to solve my problems.
I'm glad to be of help!
Thanks for the amazing videos!!
Thanks!
Thanks for the videos! Like another commenter, I can't seem to figure out how you got 0.200 at the time mark of 7:20 in your video.
Jenny FromDaBloc It's an area under the standard normal curve, and it can be found with software or a standard normal table. I have videos that show how to use the standard normal table. Cheers.
Take the Z value (at 0.842) and subtract it from 0.5. Because we want the probability above 23grams (at least) and the Z value 0.842 represents the area from 0 to 0.842 you need to subtract it from the whole area (which is 0.5 on the normal distribution table).
Very helpful, Thank you!
+JustTheRickyshow You're welcome!
@8:39 why do we choose the middle data? We have to draw a sample of size 4 so we can take four values from the left or right as well? And will the mean will be the same as 21.4 for all samples we draw from this distribution?
I don't know what you mean by "why do we choose the middle data". We're randomly picking 4 values from that distribution; they'll be whatever they'll be. It's 4 random draws from that distribution. The distribution in green is the distribution of the mean of 4 randomly picked values from the distribution in white. Yes, each of the 4 has exactly the same distribution (the distribution in white).
@@jbstatistics I mean if we have to choose randomly 4 values. Why didn't we choose the extreme 4 values of the distribution?
@@letseconomics2938 You're not understanding what's happening. We're not choosing any specific values, or letting our judgement decide which side we want to grab them from. We're randomly picking 4 values from the distribution in white. If we do that, then the distribution of the mean of those 4 randomly picked values is the green distribution. That's just how the math works out. The green distribution has a lower variance for the reasons I discuss in the earlier part of the video.
I'll just add that sure, we might end up randomly picking 4 extreme values from one tail of the distribution from which we're sampling, that will have non-zero probability of occurring. But the probability of that is very small and is reflected in the sampling distribution of the sample mean.
Great video!
Thanks!
great video, great explanations!!! thank you
These are really great!
Thanks! I'm glad you like them.
I really like your way of explaining concepts. But if in a real case scenario where we don't have mu value(suppose a very large sample) how do we even calculate the probability??
Thanks for the compliment. As you know, we don't (generally) know mu in real life. So then we can't calculate a probability this way. The primary purpose of the discussion in this video is that we later turn this problem on its head, and use a known value of X bar to say something about the unknown value of mu. Working with the mathematical concepts discussed in this video (and some others), we can come up with an appropriate formula for a confidence interval for mu, and an appropriate test statistic to use in hypothesis testing. Statistical inference is built on concepts related to the sampling distribution of the estimator of a parameter.
Edit: Which is what I allude to at the end of the video.
Hey man what if thre are 2 given sample mean like between 445 and 485?
Great explanation.
good explaining!!!!!!!!!
Thank you, very interesting
how did he get from 0.8 to 0.2 in the first example
search for statistical table for normal distribution and see 0.8 exact in the table
Thank you master
I have one doubt , x' mean is equal to population mean when we take mean of the sampling means otherwise not ? but your told that only one sample mean is equal to the population mean how tell me ?
which is the other video you derived the formula?
+e t wondering the same thing
Same
Man, I wish you was my stats professor
I'm glad I can still be of help!
how did you calculate P(z>= 1.648)=0.046?
That area can be found by using software or a standard normal table.
thankyou!
P(Z>1.648) can be found by looking up Z of 1.648 on standard normal table.
Area under the standard normal curve to the RIGHT of Z is ~0.954.
Area to the LEFT of Z is what we are looking for, so
P(Z>1.648)= 1(total area under std. norm. curve) - 0.954 = 0.046
Can someone please refer me to the video wherein Mr jbstats prove that E({x bar}) is equal to the population mean?
Is this only for for n>=30? I remember you need to use the t distribution is n
I talk about this in detail in my other videos. When sampling from a normally distributed population, the random variable Z = (X bar - mu)/(sigma/sqrt(n)) has the standard normal distribution. When sigma is replaced with the sample standard deviation S, the quantity T = (X bar - mu)/(S/sqrt(n)) has a t distribution with n-1 degrees of freedom. This is always true when we are sampling from a normally distributed population, regardless of the sample size.
super helpful thanks
You are very welcome.
I get that the SD decreases as N increase. But I'm stuck on the idea that variance increases as sample size increases. I dont get it. Please explain
I am not sure what you are asking me. The standard deviation of the sampling distribution of the sample mean is just the square root of the variance of the sampling distribution of the sample mean. The standard deviation of the sample mean and the variance of the sample mean both decrease as the sample size increases.
That's ok I realised I read my lecture slides wrong. Was confused as to why they were saying the SD and variance increased with sample size, however they were saying the opposite. Thanks for actually replying though :)
Thank you so much
You are very welcome!
Can somebody please explain how to properly estimate the population parameters: true mu and true sigma, when we don't have them?
Can I use the sample mean as a good point estimator for the population mean? Can I use the sample standard deviation as a good point estimator for the true sigma?
thank you so much!
You are very welcome Nunya!
If sigma is not given, is it possible to do any probability calculations?
In short, no. We might conceivably have an estimate of sigma, which could give us a rough estimate of the probabilities, but without knowledge of the value of sigma the exact probabilities can't be calculated.
Thank you!
im confused about the square root of n.. i didnt get the answer.. pls. help
What part were you having trouble with? What value were you getting?
perfection!
Thanks!
So Good!
thankyou sir..............
beautiful
Thanks!
wooohoooooo thank you
why is there no sound
Hi Corina. There is sound when I play the video. I'm not sure what the problem is, but I believe it's something on your end. Cheers.
Yup it was thanks. I appreciate your good work
I hope someone still sees this, but can someone explain how 0.842 becomes 2 😭
The area to the right of 0.842 under the standard normal curve is 0.200. That's found with any statistical software, such as R, or by using a standard normal table.
This is like chinese to me. God help me.
Wheres the proof
Of what part? The mean and variance? The fact that if we're sampling from a normally distributed population, then the sample mean is normally distributed? Of the central limit theorem? I derive the mean and variance of the sampling distribution here: ua-cam.com/video/7mYDHbrLEQo/v-deo.html. I don't have video proofs of the other parts yet.
i am the 1000th like.... what was the probability of that?
Really useful, thanks a lot!
+Shailendra Pandey You are very welcome!
Thanks!