Z Tests for One Mean: Introduction
An introduction to Z tests for a population mean mu. These tests are appropriate when sampling from a normally distributed population where sigma is known. I discuss the hypotheses and the underlying logic, and work through an example. I do not discuss the rejection region or p-value in an in-depth way in this video -- I discuss them in other videos, and provide links at the end of this one.
Пікірлер: 77
Finally I’ve found best source of statistic knowledge
@jbstatistics
3 ай бұрын
I've always been here :)
These videos are the perfect complimentary study guides for my grad-level stats courses. Thank you so much. I wish my tuition dollars were going to you.
@jbstatistics
6 жыл бұрын
I'm glad to hear you find my videos helpful! I'm very glad I could help.
I have a test tomorrow; today, all these statistics are finally making some sense. Thank you!!!!!!
You are awesome!!! thank you so much for these videos, they are crystal clear and make the study of statistic a real pleasure. THANK YOU!!!!
@jbstatistics
9 жыл бұрын
You are very welcome Andrea! I'm glad to be of help.
Thank you so much for your clear explanation! All your videos are of great help to my coming exams!
My goodness! I salute you.. You have made me fall in love with statistics!
You are my blessing! Thank you so much!
Great Videos. Thanks for your work!
This was very well explained. Thank you!!
Fantastic videos! Thank you.
Your explanation is clear! Much much better than the confusing textbook. I found my textbook I read is terrible no one is going to understand it .
Very organized presentation! Thank you so much.
@jbstatistics
8 жыл бұрын
+Ar Mar Bien Gregory You are very welcome!
best video i've ever seen regarding this subject.
@rsadsultanov3071
3 жыл бұрын
So have I
the best stats video regarding hypothesis testing on earth
Comic sans is terrible. But Chalkboard, which is the font used in the annotations in this video, is well known to symbolize refinement and class. I question your visual acuity and perceptiveness.
I used pnorm(-17) to get P(Z < -17) and then used the symmetry about 0 argument.
Many thanks for your clear and concise videos
@jbstatistics
6 жыл бұрын
You are very welcome!
very high quality videos thank you
Your videos are easy to understand while being in-depth. The other videos I've come across are either overcomplicated or not comprehensive enough. How to know whether we should divide by sigma or sigma x bar when finding z-value?
thank you for the videos. i have learned a lot from your videos.
@jbstatistics
8 жыл бұрын
+Mohammed Al-Bashiri You are very welcome. I'm glad to be of help!
I went through multiple videos on hypothesis testing, but your videos are best. Thank you so much. Just one doubt, If after plotting QQ plot if we find sample is not normally distributed then what would be the go forward?
Thank you!
sir ji your i love ur teaching..
Thank you for these awesome videos. I have some suggestions for the playlists and captions. 1) I think it would be a good idea to number the playlists, so that it would be easier to know which playlists should be covered first. 2) Some videos might need prior knowledge, which are already address in another video. Putting a reference to the prerequisites of such videos would be very helpful for people like me. I am watching your videos to pass a Random Variables course, and also as a prerequisite for Design of Experiments course. Thanks again.
@williamundall6988
5 жыл бұрын
if you go to "playlists" on his channel all videos are numbered and put together in different topics
AMAZING. THANK YOU.
@jbstatistics
6 жыл бұрын
You are very welcome!
could you do something on time series analysis, i.e AR Models, White Noise
Please explain me this: when you proved normality assumption by Q-Q plot, did you mean that the sample was taken from a population following a normal distribution or distribution of the sample means will follow normal distribution?
What did you type in to R to get P(Z>17)=4.1x10(-65)?
Where did you get the data to make boxplot and qqplot?
good! i like the words below videos
Could you please explain why the test statistic Z will have standard normal distribution when Null hypothesis is true? I'm not able to understand why it is so.
Thank you sir!This video is 8 years old but well organized.I am curious which software did you use at that time.I dont know if you are reading this ,but if you are please reply 😁
thank you :)
Hey! I'm watching all of your videos linearly! Don't have many doubts till now, it's amazing :D One question: If we had known the probability of selecting a tuna piece which has more than 0.40 ppm mercury content. Then we can have a random variable X = # of tuna pieces with more than 0.40 mercury content and n = 16 X = {1,if mercury content is higher than 0.40 0, if mercury content is lower than or equal 0.40 } Then, could we have inferred something from the Binomial Distribution?
@rishabhchopra6418
6 жыл бұрын
you cannot make a binomial distribution here. You don't know the probability of getting a success or a failure, here.
I love you man! Here, I said it.
Thank you so much! Even after 9 years, this video still provide so much value. May I ask, when you say that "If the null hypothesis is true, this value that we get here should simply be a random sample from the standard normal distribution", does it mean that should Z fall into the range that you showed in the boxplot diagram(of the 16 tuna), it would mean null hypothesis is true?
@jbstatistics
2 жыл бұрын
No, it doesn't mean that. I don't mean to be a jerk, but it means just what I said --- if the null hypothesis is true (and the assumptions are true), the value of the Z test statistic will be random sample from the standard normal distribution. The standard normal distribution is a probability distribution. Under H_0, the value of Z that we get will be a random sample from this distribution.
@playmakersmusic
2 жыл бұрын
@@jbstatistics No you are not a jerk and I am really appreciative of the fact that you are helping people like me who have no statistics background. I was really confused about the random sample part, but I guess it was meant to be hypothesised right? Thank you!
@jbstatistics
2 жыл бұрын
@@playmakersmusic It's a bit of an abstract concept and if you don't fully grasp it then it's not the end of the world. The major implication of that notion is that values of Z near the middle of the standard normal distribution are ordinary, everyday values to get from that distribution, so they don't provide any evidence against the null hypothesis. Values far our in the tails are unusual values to get from the standard normal distribution, so they provide evidence against it. That's oversimplified, but the gist of it.
@playmakersmusic
2 жыл бұрын
@@jbstatistics I get what you mean now! I was watching your other video where you used Pete as an example. It makes so much sense now! Thank you so much!
nice explanation I now can do my dam sw XD you're a life saver man
@jbstatistics
6 жыл бұрын
I'm glad to be of help!
Why is it the case that if H0 is true, the Z test statistic will have the standard normal distribution?
Is there a difference between hypothesis testing about a population mean and hypothesis testing about a population proportion?
@jbstatistics
6 жыл бұрын
That depends on what you mean by "difference". The basic notion is the same, and a binomial proportion is a special type of mean, but the two tests are not exactly the same.
Do you have a video on the quantile-quantile?
@jbstatistics
10 жыл бұрын
Only on the normal QQ plot, not QQ plots in general.
how do i get p(more than or equal to 17) without using the software. help please.
@jbstatistics
7 жыл бұрын
You need to use software to find this value. One could, conceivably, approximate the integral numerically by hand, but that definitely wouldn't be practical.
Thanks...
@jbstatistics
6 жыл бұрын
You are very welcome!
why not calculate sigma_x_bar (at 8:35) from the sample data?
@jbstatistics
8 жыл бұрын
+Kieran Mace You can't calculate sigma_Xbar from sample data, as sigma is a numerical characteristic of the population. If sigma is unknown, then you can estimate it with the sample standard deviation s, and we do that when we carry out a t test. But if sigma is known, as it is in this video, then we would use it rather than an estimate of it.
I have a question, what if there is no x bar?
@jbstatistics
7 жыл бұрын
X bar is the sample mean. If the sample data hasn't been collected yet, then you can't carry out a test. (But you can map out what test you would like to carry out.)
@scgaming34
7 жыл бұрын
well, in the example given to me by my teacher, there is a mean, a standard deviation and sample size, but no x bar... and the problem involves computing the probability of a type one error
look at what you did, saturday morning and first thing I did is turn on my computer to watch statistics video, could it be.. oh god am I getting to like statistics ?!
@jbstatistics
7 жыл бұрын
There's much to like. Learning statistics is a great way to start the day!
Thanks for these videos, very helpful .1 question How come here Z=(XBAR - mu)/(sigma/sqrt(n)) but in the standardizing normal random variables video Z=(XBAR - mu)/(sigma). Can someone please clear this up for me.
Z score of 17... So you're telling me there's a chance. 😏
@jbstatistics
2 ай бұрын
Always 😀
Lost myself at quantile
@jbstatistics
7 жыл бұрын
Quantiles have that effect on people.
@kzterminator
7 жыл бұрын
What's a quantile quantile chart?
@jbstatistics
7 жыл бұрын
A quantile-quantile plot allows for a visual assessment of whether two sets of data come from the same distribution, or a single data set comes from a specific theoretical distribution. In my videos, I discuss the normal quantile-quantile plot. This type of plot allows us to visually assess whether the normality assumption appears to be reasonable.
4:07 H0 isn't true. Hos ain't loyal.
comic sans! nooooooooooooooooo