Returns per Annum on Securities in Markets & Matrixes Exercise Worksheet

User Generated

ivivinyragvabin

Economics

Description

Hello,

please read the attached file. Only question 1,2,3,4 needs to be done.

I will give a tip if everything is in order :)


Unformatted Attachment Preview

AS2109 Coursework 2 Instructions: The second part of the coursework for AS2109 consists of 6 questions and you are required to answer all of them. This is a group assignment and each group should consist of no more than 4 members. Questions 2, 3 and 6 are specific to each group, whereas questions 2 and 5 are common to all groups. For the first 5 questions, you are expected to produce analytical solutions and typeset your responses with Microsoft Word or Latex. Parts of these questions may also require that you produce graph(s) to illustrate and comment on results. In such case, you are encouraged to use Excel. Deadline: 03 April 2020. Submission Method: Online via Moodle. Question 1 The returns per annum on two securities are denoted by R1 and R2 , and are modelled as follows:   R1 ∼ N 0.1, 0.22 , R2 ∼ N 0.05, 0.042 , Corr(R1 , R2 ) = 0.5 where Corr(R1 , R2 ) denotes the correlation between R1 and R2 . (i) Suppose that $1000 are invested in the minimum variance portfolio formed from the two securities. Calculate the composition of the minimum variance portfolio and state how much is invested in each security. Next, suppose that, at the beginning of the year, the balance on an investment account is $1000. The account is invested in securities 1 and 2 only, and short-selling is permitted. Suppose also that, at the end of the year, a random sum of $1000 × (1 + Q) is withdrawn from the investment account. The random variable Q is modelled as follows: Q ∼ N (0.08, 0.12 ), Corr(Q, R1 ) = 0.8, Corr(Q, R2 ) = 0.2 If x is the portfolio weight for security 1, then the end-of-year account balance is: W1 = 1000(1 + xR1 + (1 − x)R2 ) − 1000(1 + Q) (ii) Calculate the composition of the portfolio such that the end-of-year balance on the investment account has the least variance. (iii) Let m(x) = E[W1 ] and v(x) = V ar[W1 ]. (a) Derive expressions for m(x) and v(x). (b) Let x∗ be the weight of security 1 invested in the portfolio which maximises the probability of a positive end-of-year balance on the investment account. Show that: m(x∗ )v 0 (x∗ ) = 2m0 (x∗ )v(x∗ ) Question 2 (i) (a) State the no-arbitrage principle. (b) Explain why only a few arbitrageurs are required in a market to enforce the no-arbitrage condition. (ii) The returns on securities in a market are modelled using a 2-factor model: Ri = ai + bi1 I1 + bi2 I2 + ci The following data, concerning three securities, are known: Security i Ri bi1 bi2 1 12% 0.5 1.2 2 8% 1.0 0.8 3 9% 0.7 -1.0 (a) Derive an equation for the arbitrage pricing plane in Ri − bi1 − bi2 space. (b) A fourth security exists with R4 = a, b41 = b, b42 = c Explain carefully why an opportunity for arbitrage arises. (c) Construct an arbitrage portfolio from securities 1, 2, 3 and 4 (stating relevant proportions) to exploit this arbitrage portfolio. Group Group Group Group Group Group Group Group Group 20 21 22 23 24 25 26 27 28 a = 10%, b = 0.8, c = −0.2 a = 10%, b = 0.8, c = −0.2 a = 10%, b = 0.8, c = −0.2 a = 9%, b = 0.8, c = −0.2 a = 9%, b = 0.8, c = −0.2 a = 9%, b = 0.8, c = −0.2 a = 8%, b = 0.8, c = −0.2 a = 8%, b = 0.8, c = −0.2 a = 8%, b = 0.8, c = −0.2 Group Group Group Group Group Group Group Group Group Group 29 30 31 32 33 34 35 36 37 38 Table 1: Parameter Values. a = 10%, b = 0.9, c = −0.2 a = 10%, b = 0.9, c = −0.2 a = 10%, b = 0.9, c = −0.2 a = 12%, b = 0.7, c = −0.2 a = 12%, b = 0.7, c = −0.2 a = 12%, b = 0.7, c = −0.2 a = 11%, b = 0.8, c = −0.2 a = 11%, b = 0.8, c = −0.2 a = 11%, b = 0.8, c = −0.2 a = 10%, b = 0.8, c = −0.1 Question 3 (i) A model for the price Pt of a share is: Pt = P0 exp {at + bBt } where Bt denotes the standard Brownian motion and a, b are constants. (a) Calculate the expected share price at time t given that its price is 115 pence at time 2. (b) Calculate the probability that at time t the share price is greater than 118 pence, given that it is 115 pence at time 2. (c) Evaluate:  3  1 × E P2 |P1 P13 (ii) Outline the advantages of the model in part (i) above compared to the following alternative model: Pt = P0 (1 + at + bBt ) Group 20 a = 0.05, b = 0.2, t = 4 Group 29 a = 0.06, b = 0.2, t = 7 Group 21 a = 0.05, b = 0.1, t = 4 Group 30 a = 0.06, b = 0.1, t = 7 Group 22 a = 0.05, b = 0.15, t = 4 Group 31 a = 0.06, b = 0.15, t = 7 Group 23 a = 0.04, b = 0.2, t = 5 Group 32 a = 0.07, b = 0.2, t = 8 Group 24 a = 0.04, b = 0.1, t = 5 Group 33 a = 0.07, b = 0.1, t = 8 Group 25 a = 0.04, b = 0.15, t = 5 Group 34 a = 0.07, b = 0.15, t = 8 Group 26 a = 0.03, b = 0.2, t = 6 Group 35 a = 0.08, b = 0.2, t = 9 Group 27 a = 0.03, b = 0.1, t = 6 Group 36 a = 0.08, b = 0.1, t = 9 Group 28 a = 0.03, b = 0.15, t = 6 Group 37 a = 0.08, b = 0.15, t = 9 Group 38 a = 0.02, b = 0.2, t = 10 Table 2: Parameter Values. Question 4 (i) Let {Bt , t ≥ 0} denote the standard Brownian motion and let Mt := max0≤s≤t Bt . Show that the random variables |Bt |, Mt and Mt − Bt have the same distribution. (ii) Let {Bt , t ≥ 0} denote the standard Brownian motion. (a) For any 0 ≤ s < t, show that the joint distribution of (Bs , Bt ) is a bivariate normal distribution and determine the mean vector µ and covariance matrix Σ of this bivariate normal distribution. (b) Find a matrix  A= ass ast ats att   so that [Z1 , Z2 ] defined as follows has a standard bivariate normal distribution.     B Z  1 = A  s Bt Z2 (iii) If the Brownian motion {Bt , t ≥ 0} has natural filtration denoted by Ft then (a) Compute E[Bt4 |Fs ] for t > s ≥ 0 (b) Show that {Mt , t ≥ 0} such that Mt = Bt4 −6tBt2 +3t2 is a martingale adapted to the filtration Ft . Question 5 Suppose that the standard Brownian motion {Bt } is sampled at discrete points in time: t0 , t1 , t2 , . . . . The points are equi-distant, such that t1 − t0 = t2 − t1 = · · · = h, where h is a constant. Define: ej = Bt and ∆B ej = B ej − B ej−1 B j e1 − B e2 . (i) State the distribution, mean and variance of B (ii) Let Yn = Pn j=1 ej−1 ∆B ej . Show that {Yn } is a martingale with respect to the B en . (Note: You do not need to show that E|Yn | < ∞.) filtration of B (iii) Let Xn = Pn ej ∆B ej . Derive an expression for E[Xn ], and hence state whether B en . {Xn} is a martingale with respect to the filtration of B j=1 Question 6 You are considering investing in a project in the food and beverages industry. This can involve any product of your choice. After some investigation, you decide that the market is liquid enough and that the risk of the project you want to invest in can be explained by the risk of a similar (homogeneous) product or a combination of such products that are already traded in the market. In order to determine the fair rate of return for your project, you decide to put together a two-factor model. As discussed in the lectures, in terms of explanatory variables you are free to use a wide range including: i. market indices: for whole market or country-index or industry index. ii. macroeconomic factors: GDP-growth, unemployment factor, inflation, etc. iii. fundamental company-specific factors: dividend yield, company size, earnings, etc. When answering this question, make sure that you: (1) State the assumptions of your model and define clearly all the associated notation. (2) Use empirical data over a sufficiently long period of time (e.g. five years) for the explanatory variables. (3) Use Excel or any other software of your choice to estimate the values for the parameters of the regression model as well as the equation of the regression model. (4) Give a brief intuitive interpretation of the regression plane.
Purchase answer to see full attachment
User generated content is uploaded by users for the purposes of learning and should be used following Studypool's honor code & terms of service.

Explanation & Answer

Finally done xDI have also attached it in PDF incase some symbols or boxes got moved

Question 1
The returns per annum on two securities are denoted by 𝑅1 and 𝑅2 , and are modelled as follows:
𝑅1 ~𝑁(0.1,0.22 ), 𝑅2 ~𝑁(0.05,0.042 ), 𝐶𝑜𝑟𝑟(𝑅1 , 𝑅2 ) = 0.5
where 𝐶𝑜𝑟𝑟(𝑅1 , 𝑅2 ) denotes the correlation between 𝑅1 and 𝑅2
(i)

Suppose that $1,000 are invested in the minimum variance portfolio formed from the two
securities. Calculate the composition of the minimum variance portfolio and state how
much is invested in each security.

From the above information we have the following parameters:
𝐸(𝑅1 ) = 0.1, 𝜎12 = 0.22 , 𝐸(𝑅2 ) = 0.05, 𝜎22 = 0.042 , 𝐶𝑜𝑟𝑟(𝑅1 , 𝑅2 ) = 0.5
Let us define 𝑥1 as the weight of security 1, 𝑥1 that minimizes the variance is as follows:
𝑥1 𝑚𝑖𝑛 =

𝜎22 − 𝐶𝑜𝑟𝑟(𝑅1 , 𝑅2 )
0.042 − 0.5
=
= 0.5191
𝜎12 + 𝜎22 − 2𝐶𝑜𝑟𝑟(𝑅1 , 𝑅2 ) 0.22 + 0.042 − 2(0.5)

Thus, the weight of security 2, 𝑥2 = 1 − 0.5191 = 0.4809
Therefore, the composition of minimum variance portfolio is 51.91% of the investment is in security 1
and 51.91% of the investment is in security 2.
Finally, we find how much money is invested in each security,
In security 1: $ 1,000(0.5191) = $ 519.10
In security 2: $ 1,000(0.4809) = $ 480.90

Next, suppose that, at the beginning of the year, the balance on an investment account is $1,000. The
account is invested in securities 1 and 2 only, and short selling is permitted.
Suppose also that, at the end of the year, a random sum of $1000 × (1 + 𝑄) is withdrawn from the
investment account. The random variable Q is modelled as follows:
𝑄~𝑁(0.08,0.12 ), 𝐶𝑜𝑟𝑟(𝑄, 𝑅1 ) = 0.8,

𝐶𝑜𝑟𝑟(𝑄, 𝑅2 ) = 0.2

If x is the portfolio weight for security 1, then the end-of-year account balance is:
𝑊1 = 1000(1 + 𝑥𝑅1 + (1 − 𝑥)𝑅2 ) − 1000(1 + 𝑄)
(ii)

Calculate the composition of the portfolio such that the end-of-year balance on the
investment account has the least variance.

The variance of 𝑊1 is
𝑣(𝑥) = 24,000𝑥 2 − 84,000
The calculation of this is in part (iii)
We determine the derivative

𝑣 ′ (𝑥) = 48,000𝑥
Note that the value that minimizes this is x=0
Therefore, for of end-of-year balance to have the least variance, we must invest all the money
in security 2

(iii)

Let 𝑚(𝑥) = 𝔼[W1 ] and 𝑣(𝑥) = 𝑉𝑎𝑟[𝑊1 ]
(a) Derive expressions for m(x) and v(x)

First,
𝑊1 = 1000 − 1000𝑅1 𝑥 + 1000(1 − 𝑥)𝑅2 − 1000 − 1000𝑄
𝑊1 = 1000𝑅1 𝑥 + 1000(1 − 𝑥)𝑅2 − 1000𝑄
We take the expected value
𝑚(𝑥) = 𝔼[W1 ] = 𝔼[1000𝑅1 𝑥 + 1000(1 − 𝑥)𝑅2 − 1000𝑄]
= 100𝑥𝔼[R1 ] + 1000(1 − X)𝔼[𝑅2 ] − 1000𝔼[Q]
= 100𝑥(0.1) + 100(1 − 𝑥)(0.05) − 1000(0.08)
= 10𝑥 + 5(1 − 𝑥) − 80
= 5𝑥 − 75


We have 𝑚(𝑥) = 5𝑥 − 75

Next, we take the variance
𝑣(𝑥) = 𝑉𝑎𝑟(𝑊1 ) = 𝑉𝑎𝑟(1000𝑅1 𝑥) + 𝑉𝑎𝑟(1000(1 − 𝑥)𝑅2 ) − 𝑉𝑎𝑟(1000𝑄)
= 10002 𝑥 2 𝑉𝑎𝑟(𝑅1 ) + 10002 (1 − 𝑥)2 𝑉𝑎𝑟(𝑅2 ) − 10002 𝑉𝑎𝑟(𝑄)
= 10002 𝑥 2 (0.22 ) + 10002 (1 − 𝑥)2 (0.042 ) − 10002 (0.1)
= 40,000𝑥 2 + 16,000(1 − 𝑥)2 − 100,000)
= 24,000𝑥 2 − 84,000
Therefore,


𝑣(𝑥) = 24,000𝑥 2 − 84,000

(b) Show that 𝑚(𝑥 ∗ )𝑣 ′ (𝑥 ∗ ) = 2𝑚′ (𝑥 ∗ )𝑣(𝑥 ∗ )
We know that 𝑚(𝑥) = 5𝑥 − 75 and therefore 𝑚′(𝑥) = 5
𝑣(𝑥) = 24,000𝑥 2 − 84,000 and therefore 𝑣′(𝑥) = 48,000𝑥
Now, 𝑚(𝑥 ∗ )𝑣 ′ (𝑥 ∗ ) = (5𝑥 ∗ − 75)48,000𝑥 ∗ = 240,000𝑥 ∗2 − 3,600,000𝑥 ∗
2𝑚′ (𝑥 ∗ )𝑣(𝑥 ∗ ) = 2(5)( 24,000𝑥 ∗2 − 84,0002∗ ) =240,000𝑥 ∗2 − 3,600,000𝑥 ∗
Thus, 𝑚(𝑥 ∗ )𝑣 ′ (𝑥 ∗ ) = 2𝑚′ (𝑥 ∗ )𝑣(𝑥 ∗ ) are both equal to240,000𝑥 ∗2 − 3,600,000𝑥 ∗

Question 2

(i)
(a) State the no-arbitrage principle

In a financial market, there must not be any free risk profits. When arbitrage opportunities
exist, an investor can earn riskless profits and still make a net investment. Market prices will
move in such a way to rule out arbitrage opportunities.

(b) Explain why only a few arbitrageurs are required in a market to enforce the no-arbitrage
condition.

When there are arbitrage opportunities, investors will want to take a large a position as
possible to increase their earnings. Only a few arbitrageurs are required to induce the price
pressures that are needed to restore equilibrium.

(ii)

The returns on securities in a market are modelled using a 2-factor model:
𝑅𝑖 = 𝑎𝑖 + 𝑏𝑖1 𝐼1 + 𝑏𝑖2 𝐼2 + 𝑐𝑖
The following data, concerning three securities are known:
Security i
1
2
3

𝑅̅𝑖
12%
8%
9%

𝑏𝑖1
0.5
1.0
0.7

𝑏𝑖2
1.2
0.8
-1.0

(a) Derive an equation for the arbitrage pricing plane in 𝑅̅𝑖 − 𝑏𝑖1 − 𝑏12 space.
We have the three equations:
12 = 𝑎𝑖 − 0.5𝑏𝑖1 − 1.2𝑏𝑖2
8 = 𝑎𝑖 − 𝑏𝑖1 − 0.8𝑏𝑖2
9 = 𝑎𝑖 − 0.7𝑏𝑖1 + 1.2𝑏𝑖2
These are three equations with three unknowns: 𝑎𝑖 , 𝑏𝑖1 , 𝑏𝑖2
Solving the system of equations, we get the solutions:
𝑎𝑖 = 14.90,

𝑏𝑖1 = 7.45, 𝑏𝑖2 = −0.69

Thus, the equation for the arbitrage pricing plane in the 𝑅̅𝑖 − 𝑏𝑖1 − 𝑏12 space is
𝑅̅𝑖 = 14.90 + 7.45𝑏𝑖1 − 0.69𝑏12

(b) A fourth security exists with
𝑅̅4 = 10%, 𝑏41 = 0.9, 𝑏42 = −0.2
Explain carefully why an opportunity for arbitrage arises.

The new security is not the arbitrage pricing plane calculated above,
Now, let’s see if we can find weights such that:
The weighted average of the sensitivity 𝑏𝑖1 is
𝑤1 (0.5) + 𝑤1 (1) + 𝑤2 (0.7) = 0.9
The weighted average of the sensitivity 𝑏𝑖2 is
𝑤1 (1.2) + 𝑤2 (0.8) + 𝑤3 (−1) = −0.2
And we must have
𝑤1 + 𝑤2 + 𝑤3 = 1
This gives us another system of three equations with three unknowns and gives us the
solution
𝑤1 = 0
𝑤2 = 0.67
𝑤3 = 0.33
This portfolio is on the arbitrage pricing plane
Therefore, an opportunity for arbitrage arises since an investor and choose a security on the
plane that has the same risk as the fourth security with a better return.

(c) Construct an arbitrage portfolio from securities 1, 2, 3 and 4 (stating relevant proportions) to
exploit this arbitrage portfolio.

Security i
1
2
3
4

𝑅̅𝑖
12%
8%
9%
10%

𝑏𝑖1
0.5
1.0
0.7
0.9

𝑏𝑖2
1.2
0.8
-1.0
-0.2

Portfolio 4 is underpriced, and we would benefit form weighting security 4 heavily.
We can construct the portfolio such that we have the following weights (proportions)

Security i
1
2
3
4

Weight(proportions)
1/4
1/4
1/4
2/4

Question 3

(i)

A model for the price 𝑃𝑡 of a share is:
𝑃𝑡 = 𝑃0 exp {𝑎𝑡 + 𝑏𝐵𝑡 }
where 𝐵𝑡 denotes the standard Brownian motion and a, b are constants.
(a) Calculate the expected share price at time 𝑡 given that its price is 115 pence at time 2.

We have,
𝑃𝑡 = 𝑃0 exp {0.06𝑡 + 0.15𝐵...


Anonymous
Great! Studypool always delivers quality work.

Studypool
4.7
Trustpilot
4.5
Sitejabber
4.4

Related Tags