Statistics Confidence Interval Definition and formula

[Risespunk]

I swear this video explained everything so clearly in 1 min, while my book spend 20 pages beating around the bush and not giving me a clear idea of the concept. Great video! :D

[SSJChad]

Short and sweet. Well done.

[gamevirus01]

MGZ OUT HAHAH NICE ONE

[spartant22]

Thank you for making sense!!  Great vid

[mathguyzero]

The larger the confidence interval (greater %), then the wider the range between the interval values. Think of it this way: a 100% confidence interval would state that all of the sample means would fall between 0 and infinity, therefore the true population mean is within that range too... mgz out

[joe5178]

which is wider 99% or 95%..??

how can we decrease the gap between the upper and the lower limit

[Saie]

This guy is amazing.. if he was my lectrurer, i would attend all lectures and not be failing!

[metal1881]

great vid. i missed a class period and needed to learn this for a quiz. very direct and to the point. good job mgz.

[conimahoni]

nice

[AdamFreakinWest]

By using bigbeancounter for accounting and finance, and using mathguyzero for statistics, what is the point of me ever going to class? Maybe brush up on some grammar...Great video!

[WTFurBORINGme]

i love this video! it took u less than 2 min to explain what my stats teacher couldn't in 2 weeks. thank u!

[htcoolio]

thanks for this, i have a biology exam tomorrow where i need to find the confidence limits for yeast colonies and oxygen abundance and its all about this :)

[joetube1988]

Short n sweet

[TheAynushka]

how do you know when to use this one vs the p hat and q hat over n in the margin of error???

[bdeppa88]

If you repeated this process a million times you expect that the process will yield a confidence interval that covers the true population mean 95% of the time,i.e. 950,000. 95% of the sample means will from those samples will not fall between the limits you get from one confidence interval calculation.

[mc61]

LOL MGZ OUT

[vbatth]

This is the first time im doing this without taking any courses and I need help:

A company that produces detergents wants to estimate the mean amount of detergent in 64 ounce jugs. The company knows that the variance of the amount of detergent in such jugs is 0.04 ounces. If the company wants to estimate the amount of detergent within 0.03 of the true mean amount with 94% confidence, how large sample should they take?

If possible can u do this problem in a video?? Please.

[caramelcomments]

thank you so much for simplifying this! I was having a tough time understanding this basic concept. i'm not so sure about statistics in a self study environment. you have truly helped me.

[caltho12]

Thanks this is good for my Stats108 revision

[jclwhite]

Thanks for the post, very useful and simple vid

[natthetubanut]

i'm currently doing stats and have spss, thanks for the short, informative vid. although graph with the bars etc would be equally visually reinforcing aswell.

thanks for the vid

[houstontexans77449]

I have to say you are awesome!!! I looked on the internet trying to find how to find the confidence interval for about 30 minutes and still didn't understand. However, after watching your video I completely understood it in less than 1.5 minutes! AWESOME!!

[bagels75]

Thank you so much

[bagels75]

Thank you so much

[Farnam14]

god you are amazing. Thank you so much man.

[T3chMastR]

Thank you. Rock Chalk.

[T3chMastR]

Thank you. Rock Chalk.

[T3chMastR]

Thank you. Rock Chalk.

[T3chMastR]

Thank you. Rock Chalk.

[jdiv70]

Awesome! Thanks!

[IxRoachxI]

He is actually incorrect in his explanation. In repeated sampling we're not making a confidence interval that tells us the percantage of where the sample mean lies. Were making a percentage of time we capture the TRUE POPULATION MEAN(U). If it isa 95% confidence interval. 95% of the time we will capture the true value in that interval.

[dipsetmeng]

MGZ OUT!!! haha Thank you

[deefromott]

thanks. why do you refer to yourself in the third person?

[burnthosefish]

MGZ OUT!

[nikkalove]

Just a quick correction, at 1:11 you say the mean of each one of these samples should fall between those 2 levels, 99% of the time or 95% etc...

That is wrong, it should be the population mean should fall between those 2 levels 95% of the time.

[derekus111]

I thought that the confidence interval captures the the true mean of a population based on a certain level of confidence (i.e. 95%, 99%, etc.). I didn't think that the actual definition of a confidence interval was that the mean of repeated samples will fall on the interval a certain percentage of the time. Although the repeated sample means more than likely will fall on the interval, the fact that the confidence level isn't 100% is because some sample means won't fall on the interval correct?

[fornss]

I wish my stats teacher would just explain things simply like this...

[box1u]

Yes A+, these you tube vids are great tudors!

[squarcitto]

Just a question! Which is the best statistical software for you? I am using R and seems to be a very powerful package but I heard is popular only through academics... Is it true? Which are the main differences between other packages? Thanks!

[mathguyzero]

mgz uses Microsoft Excel spreadsheets for most of the basic stats computation. But if you are doing advanced stats, mgz recommends PASW (formerly SPSS) because once you figure out how to use it, this software does everything with a click or two. Good luck... mgz

[mathguyzero]

I normally dress professionally because I am in a lot of committees and meet with our school's movers and shakers frequently... mgz

[RomanByzantine]

Are you assuming that x is your mean ?

[mathguyzero]

The x-bar is the mean of a random sample. We are trying to find the true population mean with that sample and the CI formula...mgz

[RomanByzantine]

how do you know that if you repeat the sampling (whatever number of times), the mean will fall within that interval (90% or 95%,99 etc)??

[mathguyzero]

That is where advanced statistics comes in. There is a very long and complicated proof that states that this process is probably correct...mgz

[chaostik123]

@mathguyzero I'm afraid your explanation of the meaning of the CI at the end of the video is not quite accurate. The CI does not tell you that 95% (say) of hypothetical repetitions of the experiment would have a sample mean in that interval. Rather, the confidence interval is an interval which, 95% of the time it is calculated, will cover the true population mean.

Otherwise great to see stats tutorial videos being uploaded - keep up the great work.

[stikpet]

@RomanByzantine It's based on something known as the central limit theory. If you would take all possible samples out of a population and would calculate the mean for each of those samples, then those means will be normally distributed. Indeed the proof for that is rather complex.

[robbiewangyang]

@RomanByzantine law of large numbers

[trumpsnuffler]

i've just spent abotu an hour trying to teach myself this for an exam tomorrow. 1:26 minutes watching this and i've finally got it. lifesaver.

[yoyo462007]

That one minute video actually helped me tones with this. I actually understand what exactly im doing and what i need to do.

[MatteBIack]

Thank you.

[remboh]

lol hahhaa this guy explained it in about 2 minutes.... THANKS !!! youtube is the best tutor EVER !!

[illaipa]

THANK YOU MGZ you just saved me for my statistics test tomorrow!

[mathguyzero]

Statistics starts off easy but after the midterm it can get kind of crazy-hard! Remember how the z score works because you will be using it to prove or disprove the null hypothesis.

[Blueenote]

So clear, so easy, so cool. Thanks Mathguyzero. You just saved me another 3 hours of trying to figure out CI.

[3ebaby3]

thank you!!!

[djswas]

What happened to everyone's comments?

[djswas]

What happened to everyone's comments? All i see are mgz's comments!

[jaryH3]

Hi mgz, your great (:

[mathguyzero]

mgz is happy that this video helped you...mgz

[djswas]

Hi mgz. Thanks alot for this video. I should have it memorized since I have been watching it weekly. I have a question, what if you do not have the standard deviation of the population and only of the sample. Does this still work?

[trueshinobi01]

Awesome!!!!!!!!!!

[mathguyzero]

mgz is pleased that you are pleased...mgz

[ssjbriguy]

That formula is only valid if you know sigma (population std. dev). You should add to this video showing the version of the formula using T with df = n-1

[Arieona]

o.o I actually...get it now! XD Thanks!!

[aabbhfg]

its not Z* that is the critical value its just the Z. The star is only a multiply sign

[mathguyzero]

Ooopsie. You are right! Thanx for catching that...mgz

[smedheat]

thanks for writing so big in this video

[mathguyzero]

mgz is a big guy...mgz

[jesterman1013]

It should have 95% coverage, but it really doesn't with certain parameters, especially population proportions!

[peekaboo2388]

thx it really helped..because i was lost and with your video i figured it out in those two minutes :)

[mathguyzero]

That is the reason that mgz is here...mgz

[CkayeJ1978]

Thank you, thank you, thank you! Oh my god, I have been spending two days straight trying to get this, and FINALLY it made sense. Thanks for using the K.I.S.S. theory, Keep It Simple Stupid! Thank you!

[mathguyzero]

mgz is pleased...mgz

[GreenAms]

MGZ OUT---LOL --- great video you helped me alot

[addicted2utubes]

thank you!

[solitudefromwithin]

thanks for that, i am doing a stats assignment for uni and this had me stumped and i wasnt able to find what i needed elsewhere so thnk you it helped a lot :)

Jo

[pclp007]

i am cramming for my exam which takes place in 8 hours. I was stuck but this has probably saved my ass. Well done. Phew

[xxslimjimtightxx]

thanks, great job explaining hope this will help me on my unit test :)

[cooliojrm]

thank you sir!definately helped!

[mathguyzero]

Mgz is pleased that it helped...mgz

[mathguyzero]

No silly... thank Mgz!... mgz

[jrmk9]

OMG! thank God for this video :-D

[mariayaya]

thanx mgz ur the best

[mathguyzero]

Mgz is humbled by your praise...mgz

[mathguyzero]

Mgz is happy...mgz

[marianomartene]

Excelent! K-I-S Explanation

[mathguyzero]

Different brains make different paths to resolution...Is one better that the other?... Depends on the brain of the beholder... mgz

[jihadboustany]

wow my professor made this seem so difficult

[spamrog]

The formula and instructions are correct but this presenter makes a basic error in the interpretation of the confidence interval. At time 1:01 he says the mean of the different sample means will fall in the intervals. This is not true. In those millions of diff samples 95%, 99% etc will contain the true mean OF THE POPULATION. This is a common error that many students make. See The Basic Pract. of Statistics by Moore 4th edition pg 347 for more information.

[momoney26]

great post mgz

[mathguyzero]

Thanx....mgz

[tyguy7902]

man, i was sitting here for 20 minuets trying to find the answer to my problem, and then stumbled upon this video, then it took me like 15 seconds haha. great video!

[mathguyzero]

mgz is pleased... keep on mathin'...mgz

[powerinformation]

great video thanks MGZ

[djilltactics]

light bulb just appeared over my head...wow...thanks man.