CHAPTER 5 – CONTINUOUS
PROBABILITY DISTRIBUTIONS
OVERVIEW
• Probability – Density Function
• The Normal Distribution
• Properties of the Standard Normal Distribution
• Conversion from an to an
• Normal Approximation to the Binomial Distribution
More details on page 116 & 117:
Unlike the Binomial Distribution which has discrete random variables,
continuous probability distributions use continuous random variables. Another
name for a continuous probability distribution is a density curve.
Density Curve Properties:
1. Area under the curve is equal to 1. (Area and probability are similar).
2. Every point under the curve has a vertical height greater than 0.
3. Probability is found on intervals.
Area = 1
The Normal Distribution is the most common continuous probability distribution.
The Normal Distribution has the characteristic bell – shaped curve, symmetric
about the mean and the area under the curve is equal to 1.
Denoted as , where μ is the mean and is the variance.
When Standard Normal Distribution
Empirical Properties of
Standard Normal
Distribution.
More details in Section
5.4 & 5.5.
z – score:
tells us the
number of
deviations
the
measure is
away from
the mean.
Negative
z-scores
are to the Positive z-
left. scores are to
the right.
Find the areas of the
following questions using
Empirical Rule.
What happens if you need to find the areas of z-scores that are not included in the Empirical Rule?
We use the Normal Distribution tables….
More details on this Example on page 124, Example 5.14.
Thus, about 87% of children have normal lung growth.
Conversion from an distribution to an distribution.
To convert, you need to Note: Z-scores are
find the z-score that
corresponds to the value
rounded to two decimal
we want to standardize. places in order to get
proper probabilities
from the table.
Thus, about 15.58% of the population will have mild hypertension.
Note: Probabilities from the table are an estimate but using R will get a more precise answer.
Thus, 2.28% of trees from this area have an unusually large diameters.
Example: The serum cholesterol levels of 12 – to 14 – year – olds follow a normal
distribution with a mean of 155 mg/dl and standard deviation of 27 mg/dl. For the serum
cholesterol distribution, find the 80th percentile.
You can find the z – score by using a table or through technology. Technology
would be an exact z – score and the table would be in estimate.
OR
Normal Approximation to the Binomial Distribution
If n is large, the binomial distribution is very cumbersome so a Normal Approximation is easier.
When is the best time to use the Normal Approximation? When n is moderately large and p is
not too extreme, then the binomial distribution tends to be symmetric and is well approximated
by a normal distribution.
𝑋~𝑁 𝑛𝑝, 𝑛𝑝𝑞 → 𝑋~𝑁 25 0.4 , 25 0.4 0.6 → 𝑋~𝑁(10,6)
6.5 − 10 12.5 − 10
𝑃 6.5 < 𝑌 < 12.5 = 𝑃 <𝑧< = 𝑃 −1.43 < 𝑧 < 1.02
6 6
= Φ 1.02 − Φ −1.43
= 0.8461 − 0.0764 = 0.7697
𝑋~𝑁 𝑛𝑝, 𝑛𝑝𝑞 → 𝑋~𝑁 100 0.6 , 100 0.6 0.4 → 𝑋~𝑁(60,24)
49.5 − 60 75.5 − 60
𝑃 49.5 < 𝑌 < 75.5 = 𝑃 <𝑧< = 𝑃(−2.14 < 𝑧 < 3.16)
24 24
= Φ 3.16 − Φ −2.14 = 0.9992 − 0.0162 = 0.983
Thus, 98.3% will have a normal range of neutrophils.
𝑋~𝑁(60,24)
Abnormally high:
75.5 − 60
𝑃 𝑋 > 76 = 𝑃 𝑌 > 75.5 = 𝑃 𝑧 > = 𝑃 𝑧 > 3.16 = 1 − Φ 3.16 = 1 − 0.9992 = 0.0008
24
Abnormally low:
49.5 − 60
𝑃 𝑋 < 49 = 𝑃 𝑌 < 49.5 = 𝑃 𝑧 < = 𝑃 𝑧 < −2.14 = Φ −2.14 = 0.0162
24
Thus, 0.1% of people will have abnormally high neutrophil counts and 1.6% will have
abnormally low neutrophil counts.
