This is an example of good teaching; I am reviewing statistics which I studied many years ago, and this video shows me how much we do not know, even when we have studied a subject. The explanations and the illustrations aid learning. i wish teachers will watch this video to see an example of good instructions.
I learned 3 days of lecture material ( one hour and 15 mins each lecture...) in 44 minutes. You are a godsend. Also the little "pep talk" in the beginning seriously helped. Wonderful work Brandon. Thank you!
Thank you so much!!! Doing an on-line stats class for a doctorate program is a nightmare. . .you have increased my level of understanding and I am ready to view another one of your next videos.
this video just changed the game entirely! you are so encouraging and you deliver everything in such a STRAIGHT FORWARD way. I thought I was gonna have to drop this class but I'm so glad I found your videos! Thank you so much and God bless :)
I rarely comment on youtube videos, but I gotta say I'm impressed. Your explanations are thorough and concise. And your examples are unique and interesting. Thank you for your help.
Excellent teacher. I've always wanted to learn Stats, but I could never understand the meanings of the numbers. Colleges need more instructors like you.
Thank you so much!!!! I have struggled with an online course without a lecture.. You have saved me this week, and foresee myself watching them through out this term!
The videos are excellent. The fact that you pause and re-explain the preceding lines, really drills in the essence. Books dont get into so much depth and concept building. Thanks a lot!
Hey thank you so much for this video. I've been struggling in my statistics class feeling like a dumbass for more than half the semester and your introduction/conclusion really cheered me up. Please keep making more! Don't stop!
"The population parameter is either within the confidence interval or not". I like the way you emphasized that fact. I know many people who think that 95% CI means the interval which would contain the population parameter 95% of the time. This is because they get confused with the ubiquitous textbook-ish kind of explanation about CI; if you do the same kind of sampling 100 times and find the CI of each of them, 95% of the intervals will contain the population parameter . Becoming a fan of your lecture series.
Hello Brandon, love the amount of detail you get into in these videos, and the consistency of your slides is impressive. What really stands out however is the realistic examples you use! I'm tired of listening to the same old coin-toss, gum ball or whatnot problems. Much appreciated!
This has been so helpful! in less then an hour I have been trying to learn for the past three of for weeks! Just in time for finals!! Thanks so much, may God bless you!!!
That is very kind of you Marcia. :) Most teachers mean well, I am a learner first and teacher second; I think that helps! Hang in there and keep learning!
Long video, but Every part was useful. Most of us only want at most 9 minutes. This had to be done in 44. You hit the ball out of the park. The 2nd example clinched it.
I was trained in Design for Lean Six Sigma at the Black Belt level. The training was very sparse compared to the depth of the statistical subjects. For example we spent a day on hypothesis testing, Z and T distributions and how to use them. Now I am using the tools in industry and these lessons provide a great review and more in-depth understanding so that I can correctly use them.
Okay so literally I was meant to stumble across this video because i was just saying how bad I was ready to drop out and work in a factory or something because this class is so difficult I don't understand any of the concepts. So THANK YOU for the kind words. I appreciate it so much.
I'm doing my revision for CFA stats section, and I must say you are the perfect teacher! Your videos are more than enjoyable and very thorough. Looking forward to binge-watching the rest of your videos. Keep it up!
Hi John! Thanks so much! Sal Khan is one of my heroes so that was very kind of you to say. As far as the multiple samples go, actually I did not put tick marks on the sample means on purpose. I wanted to stress the concept that CIs are about mu being in or out of the sample interval. Many students turn it around and think CIs are the probability the sample mean is in the dotted interval, which is backwards and incorrect. Just my method of madness. :) All the very best, B.
Just a quick question. In the previous video you talked about the standard error of the mean. This is the standard deviation of the sample distribution. The sample distribution is ofcourse a distribution of different samples with a size of n=x. In this video however, there is just 1 sample, with n=500, which you are using to get to an estimate of the population parameter. Ofcourse the question : why aren't you taking many more samples of n=500 ? ----- edit ----- So now I think I figured it out : because we can say that with 95% confidence, all possible samples will contain the population mean (which we want to estimate) anyway, taking one sample is enough. It's the interval that 'catches' all those potential sample means. So, it's enough to take one sample and estimate from there, as long we take marginal errors into account. Correct me if I'm wrong though ...
Excellent, much better presentation than the Khan academy. One suggestion in the slide with multiple samples, I think it would make it clearer if there was a small vertical bar on each sample at the mean, to show clearly that they lie within, (or without for x5), the +/- 1.96 z values.
Thank you for your videos. Even though statistics is still a difficult class, with the reading and studying I normally do for my class, your videos clarify the concepts. This class and the next one in stats are the ones I need to complete my Bachelor's in Psychology.
U R X-cellent!!!!!!!!! U make the UNKOWN KNOWABLE. You take CONCEPTS that can be DAUNTING for MANY and You DE-MYSTIFY them. I'm Pleasantly Envious and Impressed LoL!!!👌
My stat professor is good, his only problem: He can't teach! You are a great teacher and I have most of my stats from you. My professor mentions it; you tell me what it is and show mention how to go about it. Lots of thanks.
Thanks for wonderful video. Quite knowledgeableI have two questions:1- Where that X bar (3661.5) at 35 minute comes from (in Show Rooming example)? As I tried .85*4320 to check if it is what you took but it is also 3672 and not 3661.5 as you took? Please explain. 2- I want to know which is a good book you recommend to study for Statistics (for Analytics purpose). Is anyone from below: a-Statistics for Business: Decision Making and Analysis (English) 2nd Edition Author: Dean Foster, Robert A. Stine b-Statistics for Management (English) 7th Edition Author: Richard I. Levin c- Statistics for Business and Economics (With CD) (English) 11th Edition Author: Thomas A. Williams, David R. Anderson, Dennis J. Sweeney Thanks a lot in advance.
8 hrs = 28,800 seconds 15% of 8 hrs = 4320 seconds 3661.5 is the sample mean of 125 datapoints of a sample. He didn't showed us that datapoints ...but he calculated it's mean
In Australia, about to tackle my very first bio stats class. Completely petrified but this video really help to grasp some of the concepts. Thank you..!!! Wish you could sit with me through this semester of uni, Australia is VERY nice this time of year (lol)
Thanks so much. I had been studying this from my textbook and could not wrap my head around it. In addition some things that were said in the textbook though not mentioned in your presentation suddenly made sense. I will be visiting your website often. Thanks again.
Firstly Brandon, your videos are a divine gift to me - may you be blessed. I have been teaching myself Machine Learning, Linear Algebra and Stats for 4 months. Your videos (watching them since yesterday morning) are fast-tracking me on Statistics. I understood the CONFIDENCE INTERVAL concept after watching this video painstakingly 3 times. But I found the following statement confusing - "Samples of the same size have the same standard error". This is related to the "width" you have shown around the 7 sample means. I understand that the CONFIDENCE INTERVAL is a smart concept to measure out the (1.96*SME) length on either side of the sample mean and how that allows us to construct the CONFIDENCE statement i.e. "There is a 95% probability that the population mean lies within ...". But I am unable to digest the statement "Samples of the same size have the same standard error". It also seems unnecessary. I am not sure whether saying "There exists a standard error for each sample" means anything at all. Isn't the "standard error" a metric that applies to the sampling distribution, not to a specific sample or samples?
The statement "The randomness of lies in the elements chosen for the sample" is a very critical one. It seems to me that for the concept of CONFIDENCE INTERVAL to be meaningful, the manner in which the sampling was done becomes very important. For example, in the Barnes and Noble example, the manner in which the 125 salesmen were chosen - that choice should be as random as possible. Such as repeating the following process 125 times - Select the city of the store randomly, then within that city select the shop by Annual Sales randomly, within that shop select the salesperson randomly by employee code etc. Sampling considerations such as "with or without replacement" will be important I suppose. Unless the randomness of the sample is ensured as much as possible, the confidence to make the "95% confidence" statement will be low!
Awesome video! Thank you! And wow, can't believe this video is from 2013, almost 10 years ago! Really great for this to still be useful at the present day.
First of all, your videos on statistics are mind-blowingly well structured and nicely delivered. You are an incredible teacher! I have a comment about a part of this video that I think might be misleading. At 30:33, you show the sampling mean distribution centered at 20. If that blue curve was meant to show the sampling mean distribution, it should have been centered around the population mean, not the sampling mean.
OMG!..Your presentations are very precise. I really admire and appreciate you for presenting these complex concepts in a way more comprehensible way. Thank you for saving my life Mr.Brandon Foltz.
These videos are amazingly lucid! I do have a doubt however..... At 18:32, you say “Samples of the same size have the same sample error”. But from earlier, we know the standard error to be the “Standard distribution of the sampling distribution”. Doesn’t that mean that we find the mean of samples many times over, plot their distribution and the std. of this distribution is the standard error? If that is the case, how is the standard error of a sample the same as the standard error of a distribution of samples I hope I’m able to express my query properly! Thanks!
I strongly recommended your videos to our data analysis professor at GWU, I really did not understand the concepts of the subject till I watched your videos, thank you so much for keeping it simple and interesting.
Thank you so much for these videos!! Taking an online Stat course was a HUGE mistake but you have made this so much easier to understand...I should list you as my Professor lol.
Thanks for the video, it was very helpfull. I have one question though, how were you able to derive the 3661.5 sample mean in your second example. Thanks
I am the graduate economic student and I was searching fro an specific statistical argument and just stumbled upon this video I must admit the stats that I've been studying is far more advanced than the materials he teaches however I took a pause to look at the way he teaches , Gosh is plain straight forward and far better than the teaching methods my professors used to teach , I think they liked to complicate a really easy subjects and give a bunch of non sensical denotations and terminologies where as in the real world we all will find far more practical and useful and believe it or not mathematic and stats are so sweet and appealing than the bulshits professors teach in high schools and universities
Very helpful. Do you have your videos in a particular order. It would be nice to know what you have covered before the current video so I can watch the background ones. I almost wish you have them in a course syllabus kind of order.--I know it would vary. there are are great. thanks!
Question If the standard error is the standard deviation of the sampling distribution curve The mean of the sampling distribution curve is (mean of the means) and therefore assumed to be that of the population. In your example, you get one sample data from the population. You use that mean (of the sample) as the population mean in the distribution curve. How can you do that when you have only took 1 sample population? The mean should be the mean of multiple sample populations. I understand that in a sampling distribution curve, the CI represents the likelihood of the mean being within a certain distance from the ACTUAL population mean. If we only get one sample group and find its mean, and then apply the standard error (like you did in the example), then how can the mean be true representative of the population parameter? Please help, Thanks
Hello Brandon, Your tutorials are beautifully made and simple to understand. Are these presentations available somewhere in the web? I would like to see them offline while revising the concepts.
I really enjoy and learn a lot watching your presentations, am confused on the mean of sample been 3661.5 because is not stated in the questions. thanks
hi Brandon just one question .. We calculate SEM by SD /Sqr root of sample size, its fine. But why we did not use the usual formula for Standard deviation , which is sum of squares of all deviations ?? we have means & all values, so can calculate it by using it , and then even we can found the standard deviation of the population by using SEM formula ? Can you please explain it. I watched more tha n 60 videos of you and they are so clear and this is first time I have some query .. Thanks in advance ..
I absolutely love your videos. Thanks. Can you do a video on Mathematical Statistics books that provide lots of poblems of varied difficulty to solve? That would be great.
You are doing fantastic job, Brandon. I have been following all most all of your videos. Thank you very much for your contribution. Here are my questions. Using sampling distribution, (population mean and SD is known) , we can find that any sample of same sample size, falls within the population mean and SD. In this case, based on the error, some of the samples may fall out of the boundary as well. But in this video, known population SD, and sample mean, and error that population mean will be within some interval. My questions: What if I selected wrong sample and how can I confidently say that population mean will be in so and so range? How can make sure that I have selected correct sample to say that 95% of chances that population mean will be so and so range? Please correct me If I am wrong somewhere in my understanding.
Thanks for the amazing explanation, Let's say the point estimate is the sample mean. We can repeatedly keep taking the sample means and then plot all these sample means in a histogram and we would observe a normal distribution called the sampling distribution of the sample means. The mean of this distribution would be a better estimate of the population mean and its standard deviation, called standard error would be the population standard deviation/sqrt (number of points in a sample). Won't the confidence interval(say 95%) range be (sampling distribution mean - 2 SE,sampling distribution mean + 2 SE) instead of (point estimate - 2 SE,point estimate + 2 SE)? Why would we use the sample mean(point estimate) in calculating the confidence interval range? What if that particular sample mean was like an outlier in the sampling distribution of the mean? In that case, doing +/- 2*SE wouldn't be a good judge to measure population mean right?
