Life-Cycle Portfolio Choice, the Wealth Distribution and ...
Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf ·...
Transcript of Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf ·...
![Page 1: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/1.jpg)
Growth Optimal Portfolio Selection Strategies withTransaction Cost
Laszlo Gyorfi1
Gyorgy OttucsakIstvan Vajda
1Department of Computer Science and Information TheoryBudapest University of Technology and Economics
Budapest, Hungary
September 24, 2007
e-mail: [email protected]/∼gyorfiwww.szit.bme.hu/∼oti/portfolio
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 2: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/2.jpg)
Notation
investment in the stock marketd assetss(j)n price of asset j at the end of trading period (day) n
initial price s(j)0 = 1, j = 1, . . . , d
x(j)n =
s(j)n
s(j)n−1
xn = (x(1)n , . . . , x
(d)n ) the return vector on trading period n
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 3: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/3.jpg)
Notation
investment in the stock marketd assetss(j)n price of asset j at the end of trading period (day) n
initial price s(j)0 = 1, j = 1, . . . , d
x(j)n =
s(j)n
s(j)n−1
xn = (x(1)n , . . . , x
(d)n ) the return vector on trading period n
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 4: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/4.jpg)
Portfolio selection without transaction cost
nth trading period a portfolio strategy
bn = (b(1)n , . . . , b
(d)n ) = b(x1, . . . , xn−1) = b(xn−1
1 )
b(j)n ≥ 0 gives the proportion of the investor’s capital invested in
stock j for trading period n (∑d
j=1 b(j)n = 1)
for the nth trading period, Sn−1 is the initial capital (it is invested).
Sn = Sn−1
d∑j=1
b(j)n x
(j)1 = Sn−1 〈bn , xn〉 = S0
n∏i=1
〈bi , xi 〉 = S0enWn(b)
with the average growth rate
Wn(B) =1
n
n∑i=1
log 〈bi , xi 〉 .
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 5: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/5.jpg)
Portfolio selection without transaction cost
nth trading period a portfolio strategy
bn = (b(1)n , . . . , b
(d)n ) = b(x1, . . . , xn−1) = b(xn−1
1 )
b(j)n ≥ 0 gives the proportion of the investor’s capital invested in
stock j for trading period n (∑d
j=1 b(j)n = 1)
for the nth trading period, Sn−1 is the initial capital (it is invested).
Sn = Sn−1
d∑j=1
b(j)n x
(j)1
= Sn−1 〈bn , xn〉 = S0
n∏i=1
〈bi , xi 〉 = S0enWn(b)
with the average growth rate
Wn(B) =1
n
n∑i=1
log 〈bi , xi 〉 .
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 6: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/6.jpg)
Portfolio selection without transaction cost
nth trading period a portfolio strategy
bn = (b(1)n , . . . , b
(d)n ) = b(x1, . . . , xn−1) = b(xn−1
1 )
b(j)n ≥ 0 gives the proportion of the investor’s capital invested in
stock j for trading period n (∑d
j=1 b(j)n = 1)
for the nth trading period, Sn−1 is the initial capital (it is invested).
Sn = Sn−1
d∑j=1
b(j)n x
(j)1 = Sn−1 〈bn , xn〉
= S0
n∏i=1
〈bi , xi 〉 = S0enWn(b)
with the average growth rate
Wn(B) =1
n
n∑i=1
log 〈bi , xi 〉 .
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 7: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/7.jpg)
Portfolio selection without transaction cost
nth trading period a portfolio strategy
bn = (b(1)n , . . . , b
(d)n ) = b(x1, . . . , xn−1) = b(xn−1
1 )
b(j)n ≥ 0 gives the proportion of the investor’s capital invested in
stock j for trading period n (∑d
j=1 b(j)n = 1)
for the nth trading period, Sn−1 is the initial capital (it is invested).
Sn = Sn−1
d∑j=1
b(j)n x
(j)1 = Sn−1 〈bn , xn〉 = S0
n∏i=1
〈bi , xi 〉
= S0enWn(b)
with the average growth rate
Wn(B) =1
n
n∑i=1
log 〈bi , xi 〉 .
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 8: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/8.jpg)
Portfolio selection without transaction cost
nth trading period a portfolio strategy
bn = (b(1)n , . . . , b
(d)n ) = b(x1, . . . , xn−1) = b(xn−1
1 )
b(j)n ≥ 0 gives the proportion of the investor’s capital invested in
stock j for trading period n (∑d
j=1 b(j)n = 1)
for the nth trading period, Sn−1 is the initial capital (it is invested).
Sn = Sn−1
d∑j=1
b(j)n x
(j)1 = Sn−1 〈bn , xn〉 = S0
n∏i=1
〈bi , xi 〉 = S0enWn(b)
with the average growth rate
Wn(B) =1
n
n∑i=1
log 〈bi , xi 〉 .
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 9: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/9.jpg)
1
nlog Sn ≈ 1
n
n∑i=1
E{log⟨b(Xi−1
1 ) , Xi
⟩| Xi−1
1 }
and
1
nlog S∗n ≈ 1
n
n∑i=1
E{log⟨b∗(Xi−1
1 ) , Xi
⟩| Xi−1
1 }
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 10: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/10.jpg)
1
nlog Sn ≈ 1
n
n∑i=1
E{log⟨b(Xi−1
1 ) , Xi
⟩| Xi−1
1 }
and
1
nlog S∗n ≈ 1
n
n∑i=1
E{log⟨b∗(Xi−1
1 ) , Xi
⟩| Xi−1
1 }
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 11: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/11.jpg)
Portfolio selection with transaction cost
S0 = 1, gross wealth Sn, net wealth Nn
for the nth trading period, Nn−1 is the initial capital
Sn = Nn−1 〈bn , xn〉
Calculate the transaction cost for selecting the portfolio bn+1.
Before rearranging, at the j-th asset there is b(j)n x
(j)n Nn−1 dollars.
After rearranging, we need b(j)n+1Nn dollars.
If b(j)n x
(j)n Nn−1 ≥ b
(j)n+1Nn then we have to sell and the transaction
cost at the j-th asset is
c(b
(j)n x
(j)n Nn−1 − b
(j)n+1Nn
),
otherwise we have to buy and the transaction cost at the j-th assetis
c(b
(j)n+1Nn − b
(j)n x
(j)n Nn−1
).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 12: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/12.jpg)
Portfolio selection with transaction cost
S0 = 1, gross wealth Sn, net wealth Nn
for the nth trading period, Nn−1 is the initial capital
Sn = Nn−1 〈bn , xn〉
Calculate the transaction cost for selecting the portfolio bn+1.
Before rearranging, at the j-th asset there is b(j)n x
(j)n Nn−1 dollars.
After rearranging, we need b(j)n+1Nn dollars.
If b(j)n x
(j)n Nn−1 ≥ b
(j)n+1Nn then we have to sell and the transaction
cost at the j-th asset is
c(b
(j)n x
(j)n Nn−1 − b
(j)n+1Nn
),
otherwise we have to buy and the transaction cost at the j-th assetis
c(b
(j)n+1Nn − b
(j)n x
(j)n Nn−1
).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 13: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/13.jpg)
Portfolio selection with transaction cost
S0 = 1, gross wealth Sn, net wealth Nn
for the nth trading period, Nn−1 is the initial capital
Sn = Nn−1 〈bn , xn〉
Calculate the transaction cost for selecting the portfolio bn+1.
Before rearranging, at the j-th asset there is b(j)n x
(j)n Nn−1 dollars.
After rearranging, we need b(j)n+1Nn dollars.
If b(j)n x
(j)n Nn−1 ≥ b
(j)n+1Nn then we have to sell and the transaction
cost at the j-th asset is
c(b
(j)n x
(j)n Nn−1 − b
(j)n+1Nn
),
otherwise we have to buy and the transaction cost at the j-th assetis
c(b
(j)n+1Nn − b
(j)n x
(j)n Nn−1
).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 14: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/14.jpg)
Portfolio selection with transaction cost
S0 = 1, gross wealth Sn, net wealth Nn
for the nth trading period, Nn−1 is the initial capital
Sn = Nn−1 〈bn , xn〉
Calculate the transaction cost for selecting the portfolio bn+1.
Before rearranging, at the j-th asset there is b(j)n x
(j)n Nn−1 dollars.
After rearranging, we need b(j)n+1Nn dollars.
If b(j)n x
(j)n Nn−1 ≥ b
(j)n+1Nn then we have to sell and the transaction
cost at the j-th asset is
c(b
(j)n x
(j)n Nn−1 − b
(j)n+1Nn
),
otherwise we have to buy and the transaction cost at the j-th assetis
c(b
(j)n+1Nn − b
(j)n x
(j)n Nn−1
).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 15: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/15.jpg)
Portfolio selection with transaction cost
S0 = 1, gross wealth Sn, net wealth Nn
for the nth trading period, Nn−1 is the initial capital
Sn = Nn−1 〈bn , xn〉
Calculate the transaction cost for selecting the portfolio bn+1.
Before rearranging, at the j-th asset there is b(j)n x
(j)n Nn−1 dollars.
After rearranging, we need b(j)n+1Nn dollars.
If b(j)n x
(j)n Nn−1 ≥ b
(j)n+1Nn then we have to sell and the transaction
cost at the j-th asset is
c(b
(j)n x
(j)n Nn−1 − b
(j)n+1Nn
),
otherwise we have to buy and the transaction cost at the j-th assetis
c(b
(j)n+1Nn − b
(j)n x
(j)n Nn−1
).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 16: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/16.jpg)
Portfolio selection with transaction cost
S0 = 1, gross wealth Sn, net wealth Nn
for the nth trading period, Nn−1 is the initial capital
Sn = Nn−1 〈bn , xn〉
Calculate the transaction cost for selecting the portfolio bn+1.
Before rearranging, at the j-th asset there is b(j)n x
(j)n Nn−1 dollars.
After rearranging, we need b(j)n+1Nn dollars.
If b(j)n x
(j)n Nn−1 ≥ b
(j)n+1Nn then we have to sell and the transaction
cost at the j-th asset is
c(b
(j)n x
(j)n Nn−1 − b
(j)n+1Nn
),
otherwise we have to buy and the transaction cost at the j-th assetis
c(b
(j)n+1Nn − b
(j)n x
(j)n Nn−1
).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 17: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/17.jpg)
Let x+ denote the positive part of x .
Thus,
Nn = Sn −d∑
j=1
c(b
(j)n x
(j)n Nn−1 − b
(j)n+1Nn
)+
−d∑
j=1
c(b
(j)n+1Nn − b
(j)n x
(j)n Nn−1
)+,
or equivalently
Sn = Nn + cd∑
j=1
∣∣∣b(j)n x
(j)n Nn−1 − b
(j)n+1Nn
∣∣∣ .
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 18: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/18.jpg)
Let x+ denote the positive part of x .Thus,
Nn = Sn −d∑
j=1
c(b
(j)n x
(j)n Nn−1 − b
(j)n+1Nn
)+
−d∑
j=1
c(b
(j)n+1Nn − b
(j)n x
(j)n Nn−1
)+,
or equivalently
Sn = Nn + cd∑
j=1
∣∣∣b(j)n x
(j)n Nn−1 − b
(j)n+1Nn
∣∣∣ .
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 19: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/19.jpg)
Let x+ denote the positive part of x .Thus,
Nn = Sn −d∑
j=1
c(b
(j)n x
(j)n Nn−1 − b
(j)n+1Nn
)+
−d∑
j=1
c(b
(j)n+1Nn − b
(j)n x
(j)n Nn−1
)+,
or equivalently
Sn = Nn + cd∑
j=1
∣∣∣b(j)n x
(j)n Nn−1 − b
(j)n+1Nn
∣∣∣ .
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 20: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/20.jpg)
Dividing both sides by Sn and introducing ratio
wn =Nn
Sn,
0 < wn < 1,
we get
1 = wn + cd∑
j=1
∣∣∣∣∣ b(j)n x
(j)n
〈bn , xn〉− b
(j)n+1wn
∣∣∣∣∣ .
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 21: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/21.jpg)
Dividing both sides by Sn and introducing ratio
wn =Nn
Sn,
0 < wn < 1,we get
1 = wn + cd∑
j=1
∣∣∣∣∣ b(j)n x
(j)n
〈bn , xn〉− b
(j)n+1wn
∣∣∣∣∣ .
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 22: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/22.jpg)
Sn = Nn−1〈bn , xn〉 = Sn−1wn−1〈bn , xn〉 =n∏
i=1
[w(bi−1,bi , xi−1) 〈bi , xi 〉].
Introduce the notation
g(bi−1,bi , xi−1, xi ) = log(w(bi−1,bi , xi−1) 〈bi , xi 〉),
then the average growth rate becomes
1
nlog Sn =
1
n
n∑i=1
log(w(bi−1,bi , xi−1) 〈bi , xi 〉)
=1
n
n∑i=1
g(bi−1,bi , xi−1, xi ).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 23: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/23.jpg)
Sn = Nn−1〈bn , xn〉 = Sn−1wn−1〈bn , xn〉 =n∏
i=1
[w(bi−1,bi , xi−1) 〈bi , xi 〉].
Introduce the notation
g(bi−1,bi , xi−1, xi ) = log(w(bi−1,bi , xi−1) 〈bi , xi 〉),
then the average growth rate becomes
1
nlog Sn =
1
n
n∑i=1
log(w(bi−1,bi , xi−1) 〈bi , xi 〉)
=1
n
n∑i=1
g(bi−1,bi , xi−1, xi ).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 24: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/24.jpg)
Sn = Nn−1〈bn , xn〉 = Sn−1wn−1〈bn , xn〉 =n∏
i=1
[w(bi−1,bi , xi−1) 〈bi , xi 〉].
Introduce the notation
g(bi−1,bi , xi−1, xi ) = log(w(bi−1,bi , xi−1) 〈bi , xi 〉),
then the average growth rate becomes
1
nlog Sn =
1
n
n∑i=1
log(w(bi−1,bi , xi−1) 〈bi , xi 〉)
=1
n
n∑i=1
g(bi−1,bi , xi−1, xi ).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 25: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/25.jpg)
In the sequel xi will be random variable and is denoted by Xi .Let’s use the decomposition
1
nlog Sn
=1
n
n∑i=1
E{g(bi−1,bi ,Xi−1,Xi )|Xi−11 }
+1
n
n∑i=1
(g(bi−1,bi ,Xi−1,Xi )− E{g(bi−1,bi ,Xi−1,Xi )|Xi−11 }),
therefore
1
nlog Sn ≈
1
n
n∑i=1
E{g(bi−1,bi ,Xi−1,Xi )|Xi−11 }
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 26: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/26.jpg)
In the sequel xi will be random variable and is denoted by Xi .Let’s use the decomposition
1
nlog Sn
=1
n
n∑i=1
E{g(bi−1,bi ,Xi−1,Xi )|Xi−11 }
+1
n
n∑i=1
(g(bi−1,bi ,Xi−1,Xi )− E{g(bi−1,bi ,Xi−1,Xi )|Xi−11 }),
therefore
1
nlog Sn ≈
1
n
n∑i=1
E{g(bi−1,bi ,Xi−1,Xi )|Xi−11 }
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 27: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/27.jpg)
If the market process {Xi} is a homogeneous and first orderMarkov process then
E{g(bi−1,bi ,Xi−1,Xi )|Xi−11 }
= E{log(w(bi−1,bi ,Xi−1) 〈bi , Xi 〉)|Xi−11 }
= log w(bi−1,bi ,Xi−1) + E{log 〈bi , Xi 〉 |Xi−11 }
= log w(bi−1,bi ,Xi−1) + E{log 〈bi , Xi 〉 |bi ,Xi−1}def= v(bi−1,bi ,Xi−1),
therefore the maximization of the average growth rate
1
nlog Sn
is asymptotically equivalent to the maximization of
1
n
n∑i=1
v(bi−1,bi ,Xi−1).
dynamic programming problem
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 28: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/28.jpg)
If the market process {Xi} is a homogeneous and first orderMarkov process then
E{g(bi−1,bi ,Xi−1,Xi )|Xi−11 }
= E{log(w(bi−1,bi ,Xi−1) 〈bi , Xi 〉)|Xi−11 }
= log w(bi−1,bi ,Xi−1) + E{log 〈bi , Xi 〉 |Xi−11 }
= log w(bi−1,bi ,Xi−1) + E{log 〈bi , Xi 〉 |bi ,Xi−1}def= v(bi−1,bi ,Xi−1),
therefore the maximization of the average growth rate
1
nlog Sn
is asymptotically equivalent to the maximization of
1
n
n∑i=1
v(bi−1,bi ,Xi−1).
dynamic programming problem
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 29: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/29.jpg)
If the market process {Xi} is a homogeneous and first orderMarkov process then
E{g(bi−1,bi ,Xi−1,Xi )|Xi−11 }
= E{log(w(bi−1,bi ,Xi−1) 〈bi , Xi 〉)|Xi−11 }
= log w(bi−1,bi ,Xi−1) + E{log 〈bi , Xi 〉 |Xi−11 }
= log w(bi−1,bi ,Xi−1) + E{log 〈bi , Xi 〉 |bi ,Xi−1}def= v(bi−1,bi ,Xi−1),
therefore the maximization of the average growth rate
1
nlog Sn
is asymptotically equivalent to the maximization of
1
n
n∑i=1
v(bi−1,bi ,Xi−1).
dynamic programming problem
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 30: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/30.jpg)
If the market process {Xi} is a homogeneous and first orderMarkov process then
E{g(bi−1,bi ,Xi−1,Xi )|Xi−11 }
= E{log(w(bi−1,bi ,Xi−1) 〈bi , Xi 〉)|Xi−11 }
= log w(bi−1,bi ,Xi−1) + E{log 〈bi , Xi 〉 |Xi−11 }
= log w(bi−1,bi ,Xi−1) + E{log 〈bi , Xi 〉 |bi ,Xi−1}def= v(bi−1,bi ,Xi−1),
therefore the maximization of the average growth rate
1
nlog Sn
is asymptotically equivalent to the maximization of
1
n
n∑i=1
v(bi−1,bi ,Xi−1).
dynamic programming problemGyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 31: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/31.jpg)
Algorithm 1
empirical portfolio selection
Naive approachFor the optimization, neglect the transaction costkernel based log-optimal portfolio selectionDefine an infinite array of experts B(`) = {b(`)(·)}, where ` is apositive integer.For fixed positive integer `, choose the radius r` > 0 such that
lim`→∞
r` = 0.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 32: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/32.jpg)
Algorithm 1
empirical portfolio selectionNaive approach
For the optimization, neglect the transaction costkernel based log-optimal portfolio selectionDefine an infinite array of experts B(`) = {b(`)(·)}, where ` is apositive integer.For fixed positive integer `, choose the radius r` > 0 such that
lim`→∞
r` = 0.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 33: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/33.jpg)
Algorithm 1
empirical portfolio selectionNaive approachFor the optimization, neglect the transaction cost
kernel based log-optimal portfolio selectionDefine an infinite array of experts B(`) = {b(`)(·)}, where ` is apositive integer.For fixed positive integer `, choose the radius r` > 0 such that
lim`→∞
r` = 0.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 34: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/34.jpg)
Algorithm 1
empirical portfolio selectionNaive approachFor the optimization, neglect the transaction costkernel based log-optimal portfolio selection
Define an infinite array of experts B(`) = {b(`)(·)}, where ` is apositive integer.For fixed positive integer `, choose the radius r` > 0 such that
lim`→∞
r` = 0.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 35: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/35.jpg)
Algorithm 1
empirical portfolio selectionNaive approachFor the optimization, neglect the transaction costkernel based log-optimal portfolio selectionDefine an infinite array of experts B(`) = {b(`)(·)}, where ` is apositive integer.
For fixed positive integer `, choose the radius r` > 0 such that
lim`→∞
r` = 0.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 36: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/36.jpg)
Algorithm 1
empirical portfolio selectionNaive approachFor the optimization, neglect the transaction costkernel based log-optimal portfolio selectionDefine an infinite array of experts B(`) = {b(`)(·)}, where ` is apositive integer.For fixed positive integer `, choose the radius r` > 0 such that
lim`→∞
r` = 0.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 37: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/37.jpg)
putb1 = {1/d , . . . , 1/d}
for n > 1, define the expert b(`) by
b(`)n = arg max
b∈∆d
∑{i<n:‖xi−1−xn−1‖≤r`}
ln 〈b , xi 〉 ,
if the sum is non-void,and b1 = (1/d , . . . , 1/d) otherwise, where ‖ · ‖ denotes theEuclidean norm.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 38: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/38.jpg)
putb1 = {1/d , . . . , 1/d}
for n > 1, define the expert b(`) by
b(`)n = arg max
b∈∆d
∑{i<n:‖xi−1−xn−1‖≤r`}
ln 〈b , xi 〉 ,
if the sum is non-void,
and b1 = (1/d , . . . , 1/d) otherwise, where ‖ · ‖ denotes theEuclidean norm.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 39: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/39.jpg)
putb1 = {1/d , . . . , 1/d}
for n > 1, define the expert b(`) by
b(`)n = arg max
b∈∆d
∑{i<n:‖xi−1−xn−1‖≤r`}
ln 〈b , xi 〉 ,
if the sum is non-void,and b1 = (1/d , . . . , 1/d) otherwise, where ‖ · ‖ denotes theEuclidean norm.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 40: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/40.jpg)
Aggregations: mixtures of experts
let {q`} be a probability distribution over the set of all positiveintegers `
Sn(B(`)) is the capital accumulated by the elementary strategyB(`) after n periods with an initial capital S0 = 1
after period n, aggregations with the wealths:
Sn =∑
`
q`Sn(B(`)). (1)
after period n, aggregations with the portfolios:
bn =
∑` q`Sn−1(B(`))b
(`)n∑
` q`Sn−1(B(`)). (2)
the investor’s capital is
Sn = Sn−1〈bn , xn〉w(bn−1,bn, xn−1).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 41: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/41.jpg)
Aggregations: mixtures of experts
let {q`} be a probability distribution over the set of all positiveintegers `Sn(B(`)) is the capital accumulated by the elementary strategyB(`) after n periods with an initial capital S0 = 1
after period n, aggregations with the wealths:
Sn =∑
`
q`Sn(B(`)). (1)
after period n, aggregations with the portfolios:
bn =
∑` q`Sn−1(B(`))b
(`)n∑
` q`Sn−1(B(`)). (2)
the investor’s capital is
Sn = Sn−1〈bn , xn〉w(bn−1,bn, xn−1).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 42: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/42.jpg)
Aggregations: mixtures of experts
let {q`} be a probability distribution over the set of all positiveintegers `Sn(B(`)) is the capital accumulated by the elementary strategyB(`) after n periods with an initial capital S0 = 1
after period n, aggregations with the wealths:
Sn =∑
`
q`Sn(B(`)). (1)
after period n, aggregations with the portfolios:
bn =
∑` q`Sn−1(B(`))b
(`)n∑
` q`Sn−1(B(`)). (2)
the investor’s capital is
Sn = Sn−1〈bn , xn〉w(bn−1,bn, xn−1).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 43: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/43.jpg)
Aggregations: mixtures of experts
let {q`} be a probability distribution over the set of all positiveintegers `Sn(B(`)) is the capital accumulated by the elementary strategyB(`) after n periods with an initial capital S0 = 1
after period n, aggregations with the wealths:
Sn =∑
`
q`Sn(B(`)). (1)
after period n, aggregations with the portfolios:
bn =
∑` q`Sn−1(B(`))b
(`)n∑
` q`Sn−1(B(`)). (2)
the investor’s capital is
Sn = Sn−1〈bn , xn〉w(bn−1,bn, xn−1).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 44: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/44.jpg)
Aggregations: mixtures of experts
let {q`} be a probability distribution over the set of all positiveintegers `Sn(B(`)) is the capital accumulated by the elementary strategyB(`) after n periods with an initial capital S0 = 1
after period n, aggregations with the wealths:
Sn =∑
`
q`Sn(B(`)). (1)
after period n, aggregations with the portfolios:
bn =
∑` q`Sn−1(B(`))b
(`)n∑
` q`Sn−1(B(`)). (2)
the investor’s capital is
Sn = Sn−1〈bn , xn〉w(bn−1,bn, xn−1).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 45: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/45.jpg)
Algorithm 2
empirical portfolio selection
a one-step optimization as follows:
b1 = {1/d , . . . , 1/d}
for n ≥ 1,
b(`)n = arg max
b∈∆d
∑{i<n:‖xi−1−xn−1‖≤r`}
(ln 〈b , xi 〉+ lnw(b
(`)n−1,b, xn−1)
),
if the sum is non-void,and b1 = (1/d , . . . , 1/d) otherwise.These elementary portfolios are mixed as before (1) or (2).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 46: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/46.jpg)
Algorithm 2
empirical portfolio selectiona one-step optimization as follows:
b1 = {1/d , . . . , 1/d}
for n ≥ 1,
b(`)n = arg max
b∈∆d
∑{i<n:‖xi−1−xn−1‖≤r`}
(ln 〈b , xi 〉+ lnw(b
(`)n−1,b, xn−1)
),
if the sum is non-void,and b1 = (1/d , . . . , 1/d) otherwise.These elementary portfolios are mixed as before (1) or (2).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 47: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/47.jpg)
Algorithm 2
empirical portfolio selectiona one-step optimization as follows:
b1 = {1/d , . . . , 1/d}
for n ≥ 1,
b(`)n = arg max
b∈∆d
∑{i<n:‖xi−1−xn−1‖≤r`}
(ln 〈b , xi 〉+ lnw(b
(`)n−1,b, xn−1)
),
if the sum is non-void,
and b1 = (1/d , . . . , 1/d) otherwise.These elementary portfolios are mixed as before (1) or (2).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 48: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/48.jpg)
Algorithm 2
empirical portfolio selectiona one-step optimization as follows:
b1 = {1/d , . . . , 1/d}
for n ≥ 1,
b(`)n = arg max
b∈∆d
∑{i<n:‖xi−1−xn−1‖≤r`}
(ln 〈b , xi 〉+ lnw(b
(`)n−1,b, xn−1)
),
if the sum is non-void,and b1 = (1/d , . . . , 1/d) otherwise.
These elementary portfolios are mixed as before (1) or (2).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 49: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/49.jpg)
Algorithm 2
empirical portfolio selectiona one-step optimization as follows:
b1 = {1/d , . . . , 1/d}
for n ≥ 1,
b(`)n = arg max
b∈∆d
∑{i<n:‖xi−1−xn−1‖≤r`}
(ln 〈b , xi 〉+ lnw(b
(`)n−1,b, xn−1)
),
if the sum is non-void,and b1 = (1/d , . . . , 1/d) otherwise.These elementary portfolios are mixed as before (1) or (2).
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 50: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/50.jpg)
NYSE data sets
At www.szit.bme.hu/~oti/portfolio there are two benchmarkdata set from NYSE:
The first data set consists of daily data of 36 stocks withlength 22 years (5651 trading days ending in 1985).
The second data set contains 23 stocks and has length 44years (11178 trading days ending in 2006).
Our experiment is on the second data set.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 51: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/51.jpg)
NYSE data sets
At www.szit.bme.hu/~oti/portfolio there are two benchmarkdata set from NYSE:
The first data set consists of daily data of 36 stocks withlength 22 years (5651 trading days ending in 1985).
The second data set contains 23 stocks and has length 44years (11178 trading days ending in 2006).
Our experiment is on the second data set.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 52: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/52.jpg)
Experiments on average annual yields (AAY)
Kernel based log-optimal portfolio selection with` = 1, . . . , 10
r2` = 0.0001 · d · `,
MORRIS had the best AAY, 20%
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 53: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/53.jpg)
Experiments on average annual yields (AAY)
Kernel based log-optimal portfolio selection with` = 1, . . . , 10
r2` = 0.0001 · d · `,
MORRIS had the best AAY, 20%
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 54: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/54.jpg)
The average annual yields of the individual experts and ofthe aggregations with c = 0.0015.
` c = 0 Algorithm 1 Algorithm 2
1 20% -18% -14%
2 118% -2% 25%
3 71% 14% 55%
4 103% 28% 73%
5 134% 33% 77%
6 140% 43% 92%
7 148% 37% 83%
8 132% 38% 74%
9 127% 42% 66%
10 123% 44% 62%
Aggregation with wealth (1) 137% 40% 83%Aggregation with portfolio (2) 137% 49% 89%
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 55: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/55.jpg)
Strategy 1
non-empirical strategy
0 < δ < 1 denotes a discount factordiscounted Bellman equation:
Fδ(b, x) = maxb′
{v(b,b′, x) + (1− δ)E{Fδ(b
′,X2) | X1 = x}}
.
b∗1 = {1/d , . . . , 1/d}
and
b∗i+1 = arg maxb′
{v(b∗i ,b
′,Xi ) + (1− δi )E{Fδi(b
′,Xi+1)|Xi}},
for 1 ≤ i , where 0 < δi < 1 is a discount factor such that δi ↓ 0.non-stationary policy
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 56: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/56.jpg)
Strategy 1
non-empirical strategy0 < δ < 1 denotes a discount factor
discounted Bellman equation:
Fδ(b, x) = maxb′
{v(b,b′, x) + (1− δ)E{Fδ(b
′,X2) | X1 = x}}
.
b∗1 = {1/d , . . . , 1/d}
and
b∗i+1 = arg maxb′
{v(b∗i ,b
′,Xi ) + (1− δi )E{Fδi(b
′,Xi+1)|Xi}},
for 1 ≤ i , where 0 < δi < 1 is a discount factor such that δi ↓ 0.non-stationary policy
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 57: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/57.jpg)
Strategy 1
non-empirical strategy0 < δ < 1 denotes a discount factordiscounted Bellman equation:
Fδ(b, x) = maxb′
{v(b,b′, x) + (1− δ)E{Fδ(b
′,X2) | X1 = x}}
.
b∗1 = {1/d , . . . , 1/d}
and
b∗i+1 = arg maxb′
{v(b∗i ,b
′,Xi ) + (1− δi )E{Fδi(b
′,Xi+1)|Xi}},
for 1 ≤ i , where 0 < δi < 1 is a discount factor such that δi ↓ 0.non-stationary policy
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 58: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/58.jpg)
Strategy 1
non-empirical strategy0 < δ < 1 denotes a discount factordiscounted Bellman equation:
Fδ(b, x) = maxb′
{v(b,b′, x) + (1− δ)E{Fδ(b
′,X2) | X1 = x}}
.
b∗1 = {1/d , . . . , 1/d}
and
b∗i+1 = arg maxb′
{v(b∗i ,b
′,Xi ) + (1− δi )E{Fδi(b
′,Xi+1)|Xi}},
for 1 ≤ i ,
where 0 < δi < 1 is a discount factor such that δi ↓ 0.non-stationary policy
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 59: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/59.jpg)
Strategy 1
non-empirical strategy0 < δ < 1 denotes a discount factordiscounted Bellman equation:
Fδ(b, x) = maxb′
{v(b,b′, x) + (1− δ)E{Fδ(b
′,X2) | X1 = x}}
.
b∗1 = {1/d , . . . , 1/d}
and
b∗i+1 = arg maxb′
{v(b∗i ,b
′,Xi ) + (1− δi )E{Fδi(b
′,Xi+1)|Xi}},
for 1 ≤ i , where 0 < δi < 1 is a discount factor such that δi ↓ 0.
non-stationary policy
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 60: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/60.jpg)
Strategy 1
non-empirical strategy0 < δ < 1 denotes a discount factordiscounted Bellman equation:
Fδ(b, x) = maxb′
{v(b,b′, x) + (1− δ)E{Fδ(b
′,X2) | X1 = x}}
.
b∗1 = {1/d , . . . , 1/d}
and
b∗i+1 = arg maxb′
{v(b∗i ,b
′,Xi ) + (1− δi )E{Fδi(b
′,Xi+1)|Xi}},
for 1 ≤ i , where 0 < δi < 1 is a discount factor such that δi ↓ 0.non-stationary policy
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 61: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/61.jpg)
Theorem 1
Assume
(i) that {Xi} is a homogeneous and first order Markov process,
(ii) and there exist 0 < a1 < 1 < a2 < ∞ such thata1 ≤ X (j) ≤ a2 for all j = 1, . . . , d .
Choose the discount factor δi ↓ 0 such that
(δi − δi+1)/δ2i+1 → 0
as i →∞, and∞∑
n=1
1
n2δ2n
< ∞.
Then, for Strategy 1, the portfolio {b∗i } with capital S∗n is optimalin the sense that for any portfolio strategy {bi} with capital Sn,
lim infn→∞
(1
nlog S∗n −
1
nlog Sn
)≥ 0
a.s.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 62: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/62.jpg)
Theorem 1
Assume
(i) that {Xi} is a homogeneous and first order Markov process,
(ii) and there exist 0 < a1 < 1 < a2 < ∞ such thata1 ≤ X (j) ≤ a2 for all j = 1, . . . , d .
Choose the discount factor δi ↓ 0 such that
(δi − δi+1)/δ2i+1 → 0
as i →∞, and∞∑
n=1
1
n2δ2n
< ∞.
Then, for Strategy 1, the portfolio {b∗i } with capital S∗n is optimalin the sense that for any portfolio strategy {bi} with capital Sn,
lim infn→∞
(1
nlog S∗n −
1
nlog Sn
)≥ 0
a.s.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 63: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/63.jpg)
Theorem 1
Assume
(i) that {Xi} is a homogeneous and first order Markov process,
(ii) and there exist 0 < a1 < 1 < a2 < ∞ such thata1 ≤ X (j) ≤ a2 for all j = 1, . . . , d .
Choose the discount factor δi ↓ 0 such that
(δi − δi+1)/δ2i+1 → 0
as i →∞, and∞∑
n=1
1
n2δ2n
< ∞.
Then, for Strategy 1, the portfolio {b∗i } with capital S∗n is optimalin the sense that for any portfolio strategy {bi} with capital Sn,
lim infn→∞
(1
nlog S∗n −
1
nlog Sn
)≥ 0
a.s.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 64: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/64.jpg)
Theorem 1
Assume
(i) that {Xi} is a homogeneous and first order Markov process,
(ii) and there exist 0 < a1 < 1 < a2 < ∞ such thata1 ≤ X (j) ≤ a2 for all j = 1, . . . , d .
Choose the discount factor δi ↓ 0 such that
(δi − δi+1)/δ2i+1 → 0
as i →∞, and∞∑
n=1
1
n2δ2n
< ∞.
Then, for Strategy 1, the portfolio {b∗i } with capital S∗n is optimalin the sense that for any portfolio strategy {bi} with capital Sn,
lim infn→∞
(1
nlog S∗n −
1
nlog Sn
)≥ 0
a.s.Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 65: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/65.jpg)
Strategy 2
non-empirical strategy
For any integer 1 ≤ k, put
b(k)1 = {1/d , . . . , 1/d}
and
b(k)i+1 = arg max
b′
{v(b
(k)i ,b′,Xi ) + (1− δk)E{Fδk
(b′,Xi+1)|Xi}},
for 1 ≤ i .The portfolio B(k) = {b(k)
i } is called the portfolio of expert k withcapital Sn(B(k)).Choose an arbitrary probability distribution qk > 0, and introducethe combined portfolio with its capital
Sn =∞∑
k=1
qkSn(B(k)).
stationary policy
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 66: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/66.jpg)
Strategy 2
non-empirical strategyFor any integer 1 ≤ k, put
b(k)1 = {1/d , . . . , 1/d}
and
b(k)i+1 = arg max
b′
{v(b
(k)i ,b′,Xi ) + (1− δk)E{Fδk
(b′,Xi+1)|Xi}},
for 1 ≤ i .
The portfolio B(k) = {b(k)i } is called the portfolio of expert k with
capital Sn(B(k)).Choose an arbitrary probability distribution qk > 0, and introducethe combined portfolio with its capital
Sn =∞∑
k=1
qkSn(B(k)).
stationary policy
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 67: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/67.jpg)
Strategy 2
non-empirical strategyFor any integer 1 ≤ k, put
b(k)1 = {1/d , . . . , 1/d}
and
b(k)i+1 = arg max
b′
{v(b
(k)i ,b′,Xi ) + (1− δk)E{Fδk
(b′,Xi+1)|Xi}},
for 1 ≤ i .The portfolio B(k) = {b(k)
i } is called the portfolio of expert k withcapital Sn(B(k)).
Choose an arbitrary probability distribution qk > 0, and introducethe combined portfolio with its capital
Sn =∞∑
k=1
qkSn(B(k)).
stationary policy
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 68: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/68.jpg)
Strategy 2
non-empirical strategyFor any integer 1 ≤ k, put
b(k)1 = {1/d , . . . , 1/d}
and
b(k)i+1 = arg max
b′
{v(b
(k)i ,b′,Xi ) + (1− δk)E{Fδk
(b′,Xi+1)|Xi}},
for 1 ≤ i .The portfolio B(k) = {b(k)
i } is called the portfolio of expert k withcapital Sn(B(k)).Choose an arbitrary probability distribution qk > 0, and introducethe combined portfolio with its capital
Sn =∞∑
k=1
qkSn(B(k)).
stationary policy
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 69: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/69.jpg)
Strategy 2
non-empirical strategyFor any integer 1 ≤ k, put
b(k)1 = {1/d , . . . , 1/d}
and
b(k)i+1 = arg max
b′
{v(b
(k)i ,b′,Xi ) + (1− δk)E{Fδk
(b′,Xi+1)|Xi}},
for 1 ≤ i .The portfolio B(k) = {b(k)
i } is called the portfolio of expert k withcapital Sn(B(k)).Choose an arbitrary probability distribution qk > 0, and introducethe combined portfolio with its capital
Sn =∞∑
k=1
qkSn(B(k)).
stationary policyGyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 70: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/70.jpg)
Theorem 2
Assume (i) and (ii) of Theorem 1.
Choose the discount factor δi ↓ 0 as i →∞.Then, for Strategy 2,
limn→∞
(1
nlog S∗n −
1
nlog Sn
)= 0
a.s.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 71: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/71.jpg)
Theorem 2
Assume (i) and (ii) of Theorem 1.Choose the discount factor δi ↓ 0 as i →∞.
Then, for Strategy 2,
limn→∞
(1
nlog S∗n −
1
nlog Sn
)= 0
a.s.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 72: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/72.jpg)
Theorem 2
Assume (i) and (ii) of Theorem 1.Choose the discount factor δi ↓ 0 as i →∞.Then, for Strategy 2,
limn→∞
(1
nlog S∗n −
1
nlog Sn
)= 0
a.s.
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost
![Page 73: Growth Optimal Portfolio Selection Strategies with ...oti/portfolio/lectures/costsb.pdf · Portfolio selection with transaction cost S 0 = 1, gross wealth S n, net wealth N n for](https://reader035.fdocuments.in/reader035/viewer/2022081522/5edeec33ad6a402d666a4876/html5/thumbnails/73.jpg)
Open problem
How to construct empirical (data-driven) optimal portfolioselection strategy?
Gyorfi, Ottucsak, Vajda Growth Optimal Port. Sel. Strategies with Transaction Cost