Professional Documents
Culture Documents
1 2
3 4
CONTINUOUS RANDOM
VARIABLES
Chapter 9 (Sections 9.10-9.11, 9.3), Chapter 6
1
13/10/2015
5 6
5
35
which will determine whether the
30
normal model is appropriate
4
25
• The Normal Probability Plot
Sample Quantiles
3
20
compares the z-scores with the
Frequency
values from the theoretical
15
2
distribution.
10
1
• If the Normal model is appropriate, Copyright © 2014 Pearson Canada Inc.
5
then the data falls along a straight
0
0
line – see example 0 1 2 3 4 5 -2 -1 0 1 2
x4 Theoretical Quantiles
7 8
2
13/10/2015
9 10
11 12
3
13/10/2015
13 14
1.0
lambda = 0.5
as: lambda = 1
lambda = 2
1.0
lambda = 5
0.8
$λ* .) + ≥ 0
)* (+) = ,λ-
0.8
0 .) + < 0
0.6
f(x)
0.6
f(x)
0.4
0.4
0.2
0.2
0.0
0 5 10 15 20 0 5 10 15 20
x
x
15 16
4
13/10/2015
17 18
19 20
= 1 − - $".""0# /""
• The rule for the expected value remains unchanged
• The calculation of the variance is altered
= 1 − - $".0# • We now consider the covariance between the dependent random
= 1 − 0.7788 variables X and Y:
5
13/10/2015
21 22
23 24
Summary:
Interpreting the Correlation Coefficient
Functions of a Single Random Variable
• Note that the correlation coefficient LH is bounded:
• Adding a constant (M)
−1 ≤ LH ≤ 1 • Expected value: : +M =: +M
• Variance: BC + M = BC( )
N(O(!P) N(O()
• The correlation coefficient measures the strength of the • CV: ν= =
Q(!P) Q !P
linear relationship between X and Y
• If LH > 0 then X and Y are positively linearly correlated
• Multiplying by a constant ():
• If LH < 0 then X and Y are negatively linearly correlated
• Expected value: : = :( )
• Variance: BC = 0 BC( )
N(O(() ( N(O()
• CV: ν= = =ν
Q(() (Q()
6
13/10/2015
25 26
Summary: Summary:
Functions of Two Independent Random Variables Functions of Two Dependent Random Variables
• Expected Value • Expected Value
: + D = : + :(D) : + D = : + :(D)
: M + 5D = M: + 5:(D)
: M + 5D = M: + 5:(D)
• Variance • Variance
BC + D = BC + BC(D)
BC + D = 0 BC + 0 BC(D) BC + D = BC + BC D + 2EFB( , D)
27 28
• Note that = − D
Y 20 3
can be re-expressed as:
1 + −1 D • First consider X and Y independent random variables.
• Hence using the previous notation, = 1, = −1. Then a) Let Z = 0.25X + Y. Find the standard deviation of Z:
we have U = 0.250 0 + H0 = 4.24
BC = (1)0 BC + (−1)0 BC D + 2 1 −1 COV(X, Y) b) Let Z = X – 0.5Y. Find the standard deviation of Z:
= BC + BC D − 2EFB( , D) U = 0 + (−0. 5)0 H0 = 144 + 9/4 = 12.09
7
13/10/2015
29 30
31 32
8
13/10/2015
33 34
35
Key Concepts
• For continuous random variables there are four main
types of questions: