Scribd
Upload
37 views

Example II: The Normal Distribution (A Continuous Random Variable)

The Normal distribution is the most important distribution in statistics. It is bell-shaped and symmetric, extending from negative to positive infinity. The Normal distribution is determined by two parameters: the mean (μ) and variance (σ2). Different values of these parameters produce different Normal distributions with varying means and variances. Variables that depend on many independent factors tend to be Normally distributed, such as heights, IQ scores, and sample means. The document provides examples of how to calculate probabilities and proportions for the Normal distribution using z-scores and standard Normal distribution tables.

Uploaded by

bo
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
37 views

Example II: The Normal Distribution (A Continuous Random Variable)

The Normal distribution is the most important distribution in statistics. It is bell-shaped and symmetric, extending from negative to positive infinity. The Normal distribution is determined by two parameters: the mean (μ) and variance (σ2). Different values of these parameters produce different Normal distributions with varying means and variances. Variables that depend on many independent factors tend to be Normally distributed, such as heights, IQ scores, and sample means. The document provides examples of how to calculate probabilities and proportions for the Normal distribution using z-scores and standard Normal distribution tables.

Uploaded by

bo
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Example II: The Normal distribution (a continuous random variable)

The Normal distribution is the most important in statistics. Many other distributions are
based upon or related to the Normal distribution. The Normal distribution is:

 bell shaped
 unimodal
 symmetric
 extends from - infinity to + infinity.

Its mathematical formula is:


2
x-
½ 
1   
f ( x)  P( x)  e
 2

As a shorthand we write x ~ N(, 2).

The parameters of the distribution are µ and ², like n and p for the Binomial.  turns out to
be the mean of the distribution, 2 its variance. Note that the parameters have different
meanings from the Binomial. Different values of  and 2 give rise to different Normal
distributions, changing the mean and/or variance.

Examples of variables which are Normally distributed are: IQ, height of men, height of
women, the sample mean, errors in a regression equation, etc. The Normal distribution tends
to arise when a random variable is influenced by many independent factors, none of which is
predominant. Hence height is determined by genes, diet, environment, etc. Men and women
are different because of the dominating influence of sex. Hence the height of people is not
(quite) Normally distributed.

Examples of the use of the Normal distribution

Men's heights are approximately Normally distributed with mean 174 (cm) and variance 100,
i.e. H ~ N(174, 100). What proportion of men are over 180cm tall? Note this is equivalent to
the probability of picking out a man at random from the population who is over 180cm tall.
The answer corresponds to the area under the Normal distribution, to the right of H = 180.
180

To answer this, we could find the integral of the Normal distribution formula above, but this
is difficult mathematically. Fortunately, the answers have been tabulated, once we know how
many standard deviations 180 lies above the mean. This is calculated as the z-score:

X  180  174
z   0.6
 10

Hence 180 lies 0.6 standard deviations above the mean. From tables of the standard Normal
distribution, we can then read off that 27.43% of the distribution lies above the value of z =
0.6 or, equivalently, above H = 180cm in our example.

Excerpt from table of the Standard Normal distribution

z 0.00 0.01 0.02 0.03 0.09


0.0 .5000 .4960 .4920 .4880 .. .4641
0.1 .4602 .4562 .4522 .4483 .. .4247
: : : : : .. :
0.5 .3085 .3050 .3015 .2981 .. .2776
0.6 .2743 .2709 .2676 .2643 .. .2451
0.7 .2420 .2389 .2358 .2327 .. .2148

The table gives areas under the standard Normal distribution, which has mean 0 and variance
(and standard deviation) 1, i.e. z ~ N(0,1). Thus we have established that

P(H > 180) = P(z > 0.6) = 0.2743 = 27.43%.

E.g.II: What proportion of men are less than 171cm?

171  174
z  0.3 hence 38.21% (ignore the negative sign when consulting the table)
10

E.g.III: IQ ~ N(100,100). What proportion of population have IQ between 85 and 110?

2
A B

85 100 110

Area A: z = (85 - 100)/10 = -1.5, hence 6.68% in the tail, hence A = 50% - 6.68% = 43.32%.

Area B: z = (110 - 100)/10 = 1, hence 15.87% in tail, hence B = 34.13%

Sum of areas A and B is therefore 77.45%.

You might also like