Statistical Analysis of Efficient

Portfolios
Econ 424/Amath 540
Eric Zivot
Summer 2011
Updated: August 16, 2011

Eric Zivot 2006

-0.2

-0.1

0.0

0.1

0.2

0.3

Example Data: Monthly cc Returns on

Boeing, Microsoft, Nordstrom and Starbucks

-0.4

-0.3

sbux
msft
nord
boeing

Q1

Q2 Q3
1999

Q4

Q1

Q2 Q3
2000

Q4

Q1

Q2 Q3
2001

Q4

Q1

Q2 Q3
2002

Q4

Q1

Q2 Q3
2003

Q4

Eric Zivot 2006

Full Sample CER Model Estimates

> muhat.vals
sbux
msft
nord
boeing
0.01782 -0.00006364 0.003202 0.001688
> sigmahat.vals
sbux
msft
nord boeing
0.1353 0.1375 0.1325 0.1097
> cor.mat
sbux
msft
nord
boeing

sbux
1.000000
0.295506
0.152500
0.008218

msft
0.295506
1.000000
0.383348
0.007876

nord
0.1525
0.3833
1.0000
0.2589

boeing
0.008218
0.007876
0.258940
1.000000

Eric Zivot 2006

CER Model Estimates with

Standard Errors
> rbind(muhat.vals,se.muhat)
sbux
msft
nord
boeing
muhat.vals 0.01782 -0.00006364 0.003202 0.001688
se.muhat 0.01747 0.01775446 0.017110 0.014168
> rbind(sigmahat.vals,se.sigmahat)
sbux
msft
nord boeing
sigmahat vals 0.13534
sigmahat.vals
0 13534 0.13753
0 13753 0
0.1325
1325 0
0.10975
10975
se.sigmahat 0.01236 0.01255 0.0121 0.01002

Eric Zivot 2006

0.02

0.04

95% Confidence Ellipses for and

mu

sbux

-0.0
02

0.0

boeing

0.10

0.12

0.14

0.16

sigma

Eric Zivot 2006

Global Minimum Variance Portfolio

# global minimum variance portfolio
> gmin.4 = globalMin.portfolio(er=muhat.vals,
+
cov mat=cov mat)
cov.mat=cov.mat)
> summary(gmin.4)
Call:
globalMin.portfolio(er = muhat.vals,
cov.mat = cov.mat)
Portfolio
P
tf li expected
t d return:
t
Portfolio standard deviation:
Portfolio weights:
sbux
msft
nord boeing
0.241 0.1907 0.1252 0.443

0
0.00543
00543
0.07655

Eric Zivot 2006

Global Minimum Variance Portfolio

0.2
0.0

0.1
0

Weight

0.3

0.4

Portfolio Weights

sbux

msft

nord

boeing

Assets

Eric Zivot 2006

Bootstrapping the Global Minimum Variance Portfolio

sd

sbux

4
0.0

0.02

0.04

2
1
0
0.05

0.07

0.09

0.1

0.2

0.3

0.4

Value

Value

Value

msft

nord

boeing

0.5

Density

Density

2
0.0

0.1

0.2

0.3

Value

0.4

0.5

1
0

Density

-0.02

Density

30

Density

10

10

20

20

Density

30

40

40

50

er

-0.2

0.0

0.1

0.2

Value

0.3

0.2

0.3

0.4

0.5

0.6

Value

Eric Zivot 2006

Bootstrapping the Global Minimum

Variance Portfolio
Number of Replications: 100
Summary Statistics:
Observed
Bias
er 0.00543 0.0003845
sd 0.07655 -0.0045245
sbux 0.24100 0.0177585
msft 0
0.19070
19070 -0
0.0081261
0081261
nord 0.12525 0.0015642
boeing 0.44305 -0.0111965

Mean
0.005815
0.072027
0.258759
0
0.182577
182577
0.126812
0.431852

SE
0.009668
0.009249
0.090101
0 090327
0.090327
0.089178
0.094154

Eric Zivot 2006

Efficient Portfolio: Target = 0.015

> eport.015 =
+ efficient.portfolio(er=muhat.vals,
+
cov.mat=cov.mat,
+
target.return=0.015)
> summary(eport.015)
Call:
efficient.portfolio(er = muhat.vals,
cov.mat = cov.mat, target.return = 0.015)
Portfolio expected return:
Portfolio standard deviation:
Portfolio weights:
sbux
msft
nord boeing
0.7937 -0.1673 0.1435 0.2301

0.015
0.1104

Eric Zivot 2006

Efficient Portfolio: Target = 0.015

-0.2

0.0

0.2

Weight

0.4

0.6

0.8

Portfolio Weights

sbux

msft

nord

boeing

Assets

Eric Zivot 2006

Bootstrapping the Efficient Portfolio with Target=0.015

sd

1.5
Density
D

15

0.5

10

Density
D

5
0

0.0

10
0

0.010

0.015

0.020

0.05

0.10

0.15

0.20

-0.5

0.0

0.5

Value

Value

Value

msft

nord

boeing

-0.5

0.0

0.5
Value

1.0

1.0

Density

0.0

0.0

0.5

1.0

Density

0.5

1.0
0.5
0.0

Density

1.5

1.5

1.0

1.5

0.005

2.0

0.0

1.0

50
40
30
20
0

Density
D

sbux

20

er

-1.5

-0.5

0.0

0.5

Value

1.0

0.0

0.5

1.0

Value

Eric Zivot 2006

Bootstrapping the Efficient Portfolio

with Target=0.015
Number of Replications: 100
Summary Statistics:
Observed
Bias
er
0.0150 -1.735e-017
sd
0.1104 -1.500e-002
sbux
0.7937 -4.139e-001
msft -0
0.1673
1673 2
2.781e
781e-001
001
nord
0.1435 -2.817e-002
boeing
0.2301 1.639e-001

Mean
0.01500
0.09537
0.37986
0
0.11084
11084
0.11529
0.39401

SE
0.0000
0.0293
0.2831
0 2938
0.2938
0.3653
0.3013

Eric Zivot 2006

Efficient Frontier

0.010
0.005

d
nord
boeing
0.0

Portfolio ER

0.015

sbux

msft
0.08

0.09

0.10

0.11

0.12

0.13

Portfolio SD

Eric Zivot 2006

Bootstrapping the Efficient Frontier

-0.02

0.0

0.02

Portfolio ER

0.04

10 Bootstrap Samples

0.06

0.08

0.10

0.12

0.14

Portfolio SD

Eric Zivot 2006

-0.1

0.0

0.1

0.2

0.3

0..4

0.5

24 month rolling weights in global minimum variance

portfolio: mt(n) Rolling weights in global min portfolio

-0.2

sbux
msft
nord
boeing

Q4
2000

Q1

Q2

Q3
2001

Q4

Q1

Q2

Q3
2002

Q4

Q1

Q2

Q3

Q4

2003

Eric Zivot 2006

0.04

0.06

0.08

24 month rolling means, t(n), and standard deviations,

t(n).
Rolling means and sds on global min portfolio

0.00

0.02

er
sd

Q4
2000

Q1

Q2

Q3

Q4

Q1

Q2

2001

Q3

Q4

Q1

Q2

2002

Q3

Q4

2003

Eric Zivot 2006

24 month rolling weights in efficient portfolio with target=0.015: xt(n)

-0
0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

Rolling weights in efficient portfolio with target=0.015

-0.8

-0.6

sbux
msft
nord
boeing

Q4
2000

Q1

Q2

Q3
2001

Q4

Q1

Q2

Q3
2002

Q4

Q1

Q2

Q3

Q4

2003

Eric Zivot 2006

24 month rolling means, t(n), and standard deviations, t(n).

0.04

0.06

0.08

0.10

0.12

0.14

0.16

Rolling means and sds on efficient portfolio with target=0.015

0.02

er
sd

Q4
2000

Q1

Q2

Q3
2001

Q4

Q1

Q2

Q3
2002

Q4

Q1

Q2

Q3

Q4

2003

Eric Zivot 2006

10