Confidence Interval 1

[kj369kj]

Thanks, really appreciate it when someone can clearly and energetically explain a topic to me.

[drbandar]

what is a 90% confidence interval of the mean of the population?

i need the answer , please

thanx in advance

[plan101]

thank you for sharing. This is a logic curve for me thank you!

[drjbk]

I think that's pretty neat too.

[szesamltss]

Anyone who is confuse why 40 isn't the SD of the sample, look at @davvigtu and @blacc 's explanation. Thank you to both. =)

[perriperripanpan]

Is the 'standard deviation of the sample distribution of the sample mean' the same this as the 'standard error of the mean'?

[oluwatoba11]

@ Lad812, he doesn't tell you the distribution is a normal one. So, t-distribution would have been the best (because of the estimate) if the distribution were normal. However, CLT allows us to use Z-distribution if the sample size is large enough: say 40. Although the n here is 36, it was better with "Z" than "T" distribution because you can't assume the population to be normal except stated

[ghoreyshi]

thanx brother.

[lad812]

You should be using a t distribution because you are estimating the population standard deviation with the sample standard deviation.

[wellylights]

This is awesome! thank you :)

[saan617]

Great video man..Will surely help me for my xams.. :D

[kapumizoki]

these videos are so helpful!!! 10 minutes of your video explains better than 100 pages in my stats book~~~ thankyou so much!!!

[gershonification]

Hi there, to reiterate everyone else, great video!!! Can you help me to answer this question? -Among females in Belgium between the ages of of 18 and 74 years, diastolic blood pressure is normally distributed with a mean=77mmHg, and a standard deviation of=11.6mmHg. Question: What is the probability that a randomly selected woman has a blood pressure of <60mmHg? Do i have to divide the mean by the standard deviation? Im could really use your help?

[oozecandy]

Great vid- so I'm guessing the only place the 200,000 implicitely comes into play is when we assume the population and sample mean are approximately the same?

[devilzluv]

if we're using the sample standard deviation ("40 gram sample standard deviation" ) shouldn't we be using the t-based confidence intervals.

I thought we're only meant to use z-based confidence interval only if population s.d. is known??

[BITpinched]

@devilzluv we are allowed to use z-test if the sample size is bigger than 30 too! (n>30) we can get the approximate result using z-test.

and if you noticed, T-TABLE usually only consists df number up to 30 and for this case the sample size is 36, so is kind of hard to get approximate result using t-test here.

[blaccc1]

I'm confused. isn't 40 the standard deviation of the sample? shouldn't the standard deviation of the population be 40 times 6?

[agathamit]

@blaccc1 I have the same question to Sal

[BITpinched]

@blaccc1 yes, 40 is the standard deviation of the SAMPLE.

But NOT the standard deviation of the sampling distribution of the SAMPLE MEAN! (so you cant just 40 times 6 because is different thing! )

There is a difference between SAMPLE and SAMPLE MEANS.

Try wiki this "sampling distribution of the sample mean", they will explain more about it. I'm not very good at explaining, so hope it helps =/

[Absinthus]

@BITpinched Much appreciated.

[V2PRC]

@blaccc1 Thanks for raising the question, I was confused at one moment, but then I sort of figured it out. Sal is assuming the SD of the sample is representative of the SD of the population. But what we are going after here is the SD of sample mean rather than the SD of the population.

[davvigtu]

@blaccc1 40 is the standard deviation of the weight of the apples that you checked. It's the "sample" standard deviation. If you'd done 100 different trials of 36 each, then the standard deviation of the means of each trial would be (close to) the standard deviation of the sampling distribution. But you only have one trial - of 36 samples.

Or in shorter terms: the sample deviation is "between" every apple you tested. The std dev of the sampling distribution is "between" each trial.

[FoodTech41]

@davvigtu Thanks for your explanation. I think it makes it all clear.

[shivkalra1]

@blaccc1 You are confusing between the "single sample mean" by "mean of the sample mean distribution".

[dokhterpurdyla]

@blaccc1 Hi blaccc1. 40 is the sample standard deviation (the best estimate of the population standard deviation calculated from the sample) whereas 40/6 is the standard deviation of the distribution of lots of samples of 36 apples. The population standard deviation measures the spread of all 200,000 apples away from the mean but the sampling distribution standard deviation measures the spread of the sample mean (of lots of samples) away from the population mean. Hope this helps.

[Hateusernamearentu]

@blaccc1 ..there is a formula about it..

[afaqkhanpwr]

You are the great man and greater teacher. Thanks in tons

[afaqkhanpwr]

cant we have better estimates of the population mean and standard deviation without getting mean of the many sample means and standard deviation of the mean of the sample means? Whatever Mr. Khan, thanks for your valuable efforts and time

[afaqkhanpwr]

Its not weird that khan has got such low views for this video, lols. I have watched it again and again and again and yet could get only 20% of it. May be i m duffer or perhaps the topic is too complicated. I dont understand what is the point in taking repeatedly samples of 36? To have normal distribution - but yet why we even need to have a normal distribution?

[203132]

nice, i have an exam on wednesday about this. epic

[aslwo]

@203132 How did it work out? ; D

[wookiemaster73]

first