The Student Room Group

normal distribution as question

attempted a question and the answer given is oddly not what i got even though I went over the question again with a revision book.
question is:
x~n(20,12)
find the value of a such P(X < a) = 0.40
so I converted 0.40 to 0.6554 using the normal distribution table.
so the equation will be:
(a-20/(root12)) = 0.6554
so it will simplify to a*(root12)+20 which would give A.
The answer is actually 19.1....
even if I did something wrong I do not see how this is the right answer as you have to plus 20 so it will always be bigger than 19.1... followed the revision guide and still cant get it so I was hoping the wonderful community of tsr could help me? :smile:
Original post by differentiation
attempted a question and the answer given is oddly not what i got even though I went over the question again with a revision book.
question is:
x~n(20,12)
find the value of a such P(X < a) = 0.40
so I converted 0.40 to 0.6554 using the normal distribution table.
so the equation will be:
(a-20/(root12)) = 0.6554
so it will simplify to a*(root12)+20 which would give A.
The answer is actually 19.1....
even if I did something wrong I do not see how this is the right answer as you have to plus 20 so it will always be bigger than 19.1... followed the revision guide and still cant get it so I was hoping the wonderful community of tsr could help me? :smile:


You need to use the normal distribution tables in reverse.

Usually, given a z value (no. of standard deviations), you look up the probabilty that X <= to that value in the body of the table.

Here you've been given the value in the body of the table - except the body of the table usually start at 0.5 going up so, you're going to have to find 1-0.4 in the body of the table and then it's minus the z value that gives that probability that you want.
Thanks for the help!! I have a terrible maths teacher who often jumps to the equations before explaining what each variable means which is obviously very very stupid and annoying. So just to clarify, if you don't mind me asking, what is Z, z, a or a and b and what does the formula x-mean/standard deviation actually work out and if you don't me asking as well, why does the equation do that? Revision book is crap for this topic for some reason..
z=(x - mu)/ sigma


Posted from TSR Mobile
Original post by differentiation
Thanks for the help!! I have a terrible maths teacher who often jumps to the equations before explaining what each variable means which is obviously very very stupid and annoying. So just to clarify, if you don't mind me asking, what is Z, z, a or a and b and what does the formula x-mean/standard deviation actually work out and if you don't me asking as well, why does the equation do that? Revision book is crap for this topic for some reason..


Z represent the random variable of the standardised normal distribution N(0,1), which you can find in tables.

z refers to values it takes.

There's no specific meaning to a,b in reference to the normal distribution.

If X is a random variable with distribution N(μ,σ2)N(\mu,\sigma^2) then Xuσ\dfrac{X-u}{\sigma} has distribution N(0,1), and so can be looked up in tables.
Original post by ghostwalker
Z represent the random variable of the standardised normal distribution N(0,1), which you can find in tables.

z refers to values it takes.

There's no specific meaning to a,b in reference to the normal distribution.

If X is a random variable with distribution N(μ,σ2)N(\mu,\sigma^2) then Xuσ\dfrac{X-u}{\sigma} has distribution N(0,1), and so can be looked up in tables.


Oh thanks :smile: Would you be able to write out the equation for a if you don't mind, I learn better from examples to see how it works.
Original post by differentiation
Oh thanks :smile: Would you be able to write out the equation for a if you don't mind, I learn better from examples to see how it works.


P(X<a)=0.4P(X< a ) = 0.4

Converting to the standard normal distribution:

P(Z<a2012)=0.4P(Z< \frac{a-20}{\sqrt{12}}) = 0.4

Now the cumulative function for the standard normal distribution is refered to as ϕ\phi, so we're looking to solve:

ϕ(a2012)=0.4\operatorname{\phi}(\frac{a-20}{\sqrt{12}}) = 0.4

Hence a2012=ϕ1(0.4)\frac{a-20}{\sqrt{12}} = \operatorname{\phi}^{-1}(0.4)

Now the inverse function is not usually tabled for value less than 0.5, so we exploit the symmetry of the normal distribution, and can say:

a2012=ϕ1(10.4))\frac{a-20}{\sqrt{12}} = - \operatorname{\phi}^{-1}(1-0.4))

a2012=0.253\frac{a-20}{\sqrt{12}} = - 0.253
(edited 9 years ago)

Quick Reply

Latest