Scribd
Upload
7 views

Normal Distribution

Chapter 3 covers normal distribution, emphasizing the 68-95-99.7 rule, z-values, and the relationship between mean and median. It discusses using normal distribution to approximate binomial distribution, continuity corrections, and hypothesis testing. The chapter also includes practical exercises and examples to apply these concepts effectively.

Uploaded by

abyansaami10
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
7 views

Normal Distribution

Chapter 3 covers normal distribution, emphasizing the 68-95-99.7 rule, z-values, and the relationship between mean and median. It discusses using normal distribution to approximate binomial distribution, continuity corrections, and hypothesis testing. The chapter also includes practical exercises and examples to apply these concepts effectively.

Uploaded by

abyansaami10
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 14

A2 Chapter 3: Normal Distribution

We need to know the general graph, the area, the points of inflection and the 68-95-99.7 rule.

Apply the ‘68-95-99.7 rule but with different numbers’ rule!


Do it on a calculator! We use normal, and we use CD, never PD.
If you can do these, you can do them all. And don’t worry about greater than or less than signs,
we ignore those.

Normal distribution to find the chance of being eligible, then binomial distribution to find the
chance of this many people being eligible!
Going from probability to value!

Using that to find the top x% or bottom x% and the interquartile range!

For part c, the median is always equal to the mean when we assume normal distribution!
For part a, we do the first area as 0.3, then the second area as 0.8, it’s clear when you draw a
diagram!

We need to know the formula for z-value, the Z-N graph, and what the phi notation means.
An example that uses the phi and 1-phi!

Using the z language and symmetry to write alternative expressions!


And when using our calculators, appreciate that the mean is 0 and the left side has negative
values, so a ‘really small value’ would have to be something like -10000, -100000 etc.

Doing it with numbers!

For part b, be careful as you have to convert both the lower bound and the upper bound!

Find missing mean or missing standard deviation! And yes, this uses the z language!
For this one, you could draw a diagram and use symmetry to find the mean, then substitute that
into the z method to find the standard deviation!
Be able to do these too!

Exercise 3E challenge!

Next, we can use the normal distribution to approximate a binomial distribution!


We need to know where it comes from, the two formulae, the fact that we can use them when p
is around 0.5 (binomial distribution is symmetric like the normal distribution) and n is large, and
continuity corrections.

Typo on the greater than example. For greater than, we increase it so 5 goes to 5.5.

An example!
Continuity correction problem, not fully sure about it.
Imagine we have ten children aged 0 to 9, the mean is 4.5. If you took a sample of 4 children,
the mean could be anything depending on which 4 kids you took. If you took a sample of 6
children, again the mean could be anything. Let’s say you took several samples of 4, then
calculated the mean for each sample, then plotted the means on a distribution curve, you’d get
a normal distribution. Same if you did a sample of 6.
The mean of this curve is, as you might expect, the mean of the original whole sample.
One thing we must know is that the greater the sample size, the less variation and thus the
higher the central peak.

Now we have hypothesis testing for sample sizes - be able to do these two examples! It is just:
use the formula, find the probability of the mean being above or below the given value
depending on what the sign of H1 is, then if your answer is above the significance level you say
there’s not enough evidence to reject H0. If it’s two tailed then we do both tails and remember to
halve the significance level before checking if we’re above the significance level and thus
cannot reject H0.
If they don’t give you the mean, then you say that p(x is greater or less than a, depending on
what the sign is in H1) is below the significance level, then draw a diagram and use the inverse
normal distribution on your calculator. Remember for a two-tailed test to do both sides and to
halve the significance level before checking that you’re below it to reject H0.

p-value again! We find the probability of p(mean<8.95) then double it to get the p-value.
Be careful on part b because we previously found the probability that a bolt can be sold but now
we’re asked to find the probability that a bolt cannot be sold, so we do 1 minus our previous
probability before putting it into the binomial!

Two examples with conditional probability! They’re easy though.

Ok, be careful on this one. We spotted that 127 + 6 = 133 hours lasting time, but the fact that
we’ve survived 127 hours already means we have a ‘given that’, so we have to do p(x>133)
divided by p(x>127).

You might also like