P. 1
Continuous Distributions

# Continuous Distributions

|Views: 24|Likes:

See more
See less

07/21/2010

pdf

text

original

# Continuous Distributions

 Prof G R C Nair

Objectives
 Learn the characteristics and application of the normal distribution.  Define and calculate the standard normal distribution values.  Determine the probability of an observation to be above or below a given value using the standard normal distribution.  Approximating normal distribution for binomial distributions  Learn another continuous distributions, ¶t·

Normal Distribution
 The Normal curve is continuous distribution  It is a bell-shaped curve and has a single peak at the exact center of the distribution. The arithmetic mean, median, and mode of the curve are the same and at the peak

The normal distribution is perfectly symmetrical about its mean.
Thus half the area under the curve is above the mean and half is below it.

The normal probability distribution is asymptotic. That is the curve gets closer and closer to the X-axis but never actually touches it.

l

i

:

,

Characteristics of a Normal Distribution

Normal curve is symmetrical

Mean, median, and mode are equal
x

Theoretically, curve extends to infinity

The Standard Normal Distribution
The standard normal distribution is a normal distribution with a mean of 0 and a standard deviation of 1. It is also called the z distribution.
A z-value is the distance between a selected value, designated X, and the population mean, divided by the population standard deviation. The formula is:

z !

X 

Q W

 The monthly starting salaries of recent MBA graduates follows the normal distribution with a mean of Rs 20,000 and a standard deviation of Rs 2000. What is the z-value for a salary of Rs 24,000?
X  Q z ! W 24, 000  20 , 000 ! 2000

! 2 . 00

What is the z-value of Rs 17,000?
X  Q z ! W 1 7,000  20, 000 ! 2000

!  1 . 50

A z-value of 2 indicates that the value of Rs 24,000 is two standard deviation above the mean of Rs 20,000. A zvalue of ²1.50 indicates that Rs17,000 is 1.5 standard deviation below the mean of Rs 20,000.

Areas Under the Normal Curve
 About 68 percent of the area under the normal curve is within one standard deviation of the mean. About 95 percent is within two standard deviations of the mean. Practically all (99.7%) is within three standard deviations of the mean. The total area under the curve is 1

i

,

Areas Under the Normal Curve

Q 2
x

Q
Q 1

Q2

Q3

Q 1

Q3

Example The daily water usage per person in New Delhi is normally distributed with a mean of 20 gallons and a standard deviation of 5 gallons. How many gallons of water will be used by about 68 percent of those living in New Delhi?  About 68% of the daily water usage will lie between 15 and 25 gallons.

 What is the probability that a person from New Delhi selected at random will use between 20 and 24 gallons per day?

z !

X 

Q W

!

20 

20 5

! 0 . 00

z !

X 

Q W

!

24 

20 5

! 0 . 80

 The area under a normal curve between a z-value of 0 and a zvalue of 0.80 is 0.2881.(See table)  We conclude that 28.81 percent of the residents use between 20 and 24 gallons of water per day.

Standard Normal table

r

l

i

:

,

P(0<z<.8) =.2881

0<x<.8

-4

-3

-2 -1

x

0

1

2

3

4

 What percent of the population use between 18 and 26 gallons per day?

z !

X 

Q W  Q W

!

18 

20 5  20 5

!  0 . 40

z !

X

!

26

! 1 . 20

 The area associated with a z-value of ² 0.40 is .1554.  The area associated with a z-value of 1.20 is .3849.  Adding these areas, the result is .5403.  We conclude that 54.03 percent of the residents use between 18 and 26 gallons of water per day.

Example

 The distribution of marks in a class was normal with 45 as average and 10 as std deviation. How many can be expected to have a first class (above 60%) ? How many have distinction ( above 75%) ? How many failed (below 35%) ?

Example - HW
 A corporation has 10,000 Bulbs on a city street. Life of bulb is normally distributed with 60 days mean and 20 days std deviation. How many bulbs are expected to be replaced with in i. 20 days ii. 35 days iii. 75 days , iv. between 75 and 100 days?  ans : i.228, ii. 1056, iii.7734 iv.2038

Normal approximation for Binomial
Normal distribution can be used as approximation for Binomial for large values of n. Both np and nq should be at least 5.  Take Q = np ; and Variance = npq with continuity correction factor for converting discrete to continuous by adding/subtracting, (as the case) 0.5. eg P(>10) in Binomial = P (>10.5) in Normal P(<6) in Binomial = P (< 5.5) in Normal P(5,6,7or 8) in Binomial = P(4.5 to 8.5) etc

Example
 In 10 toss of a coin , find the probability of getting head 5, 6, 7, or 8 times by normal approximation.  p=0.5, q=0.5 n=1 - So np=5, nq= 5. So Normal Approximation is just permissible.  Take Q as np =5 and W as root of npq=1.58 and take the area between 4.5 to 8.5 from std normal table. ie z =-.3165 to 2.21= .6119  (Binomial gives .6123).

¶t· Distribution
      Symmetrical and continuous. Similar to std normal curve. Lower peak and higher tails. Its center is at zero. Has one more parameter d.f.=n-1 Used for study of small samples (<30 elements).

 t = (X-Q) / (s/root n)

Assignment
 Page 271,279 Levin

scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->