Scribd
Upload
14 views3 pages

Assignment 3

The document outlines the assignment instructions for ECO3410, including mandatory and optional questions related to the expectations hypothesis, yield curves, and stock-bond correlation. Students are required to submit their answers by April 20, 2025, and can earn bonus points for correctly answering the bonus question. The assignment includes theoretical proofs, data analysis tasks, and a bonus section on consumption-based asset pricing models.

Uploaded by

luche
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
14 views3 pages

Assignment 3

The document outlines the assignment instructions for ECO3410, including mandatory and optional questions related to the expectations hypothesis, yield curves, and stock-bond correlation. Students are required to submit their answers by April 20, 2025, and can earn bonus points for correctly answering the bonus question. The assignment includes theoretical proofs, data analysis tasks, and a bonus section on consumption-based asset pricing models.

Uploaded by

luche
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

2025 Spring Instructor: Zehao Li

ECO3410 Assignment 3
Instruction:

• Everybody should answer all Mandatory Questions. The Bonus Question is optional. If you
answer the Bonus Question correctly, you receive 0.5 bonus points for your final grade.

• Please submit your answers via Blackboard. The deadline is 23:59 pm, April 20. If you miss
the deadline, you will receive zero points.

1 Mandatory Questions
1. Expectations Hypothesis

(a) Prove: under the strong[ expectations hypothesis, ] the log forward rate 𝑓𝑡(𝑛,𝑚) can be
written as 𝑓𝑡(𝑛,𝑚) = 𝑚1 E𝑡 𝑦 𝑡+𝑛
(1) (1)
+ · · · + 𝑦 𝑡+𝑛+𝑚−1 .
(b) A famous test of the EH is Campbell, J. Y., and R. J. Shiller. (1991): Yield Spreads
and Interest Rate Movements: A Birds Eye View, The Review of Economic Studies,
58, 495514. The key to their test is, under the EH,
[ ] 𝑚 ( (𝑛) )
(𝑛−𝑚) (𝑛) (𝑚)
E𝑡 𝑦 𝑡+𝑚 − 𝑦 𝑡 = 𝑦 − 𝑦𝑡 , 𝑛 > 𝑚. (1)
𝑛−𝑚 𝑡
Prove the equation. What does the equation imply about a regression test of the EH?
(c) Suppose the short-term interest rate follows an AR(1) process: 𝑦 𝑡(1) = 𝜇 + 𝜌𝑦 𝑡−1
(1)
+ 𝜀𝑡 ,
𝜌 ∈ (0, 1). Under the expectations hypothesis, what should the yield curve look like?
According to this model, what patterns of yield curve dynamics indicate violations of
the expectations hypothesis?

2. Yield Curves. Please go to the following website to access historical daily yields on US
government debt of different maturities. Download the Nominal Yield Curve data. The
log zero-coupon yields are denoted by “SVENYxx”, where xx corresponds to a particular
maturity (years). All yields are annualized percentage rates.

(a) Create a table showing the yields on Treasury securities with maturities of 1, 2, 3, 5, 7,
10, and 20-year maturities on the following specific dates:
i. December 31, 1993
ii. April 20, 1995

1
2025 Spring Instructor: Zehao Li

iii. December 27, 2000


iv. April 23, 2004
v. January 3, 2007
vi. January 3, 2014
(b) Create a plot of the yield curve for each of these dates. What does the yield curve
normally look like? What are a couple of dates where the yield curve looks different?
(c) For each of the given dates, use data on the 1, 2, and 3-year maturity yields to infer
𝑓𝑡(1,1) and 𝑓𝑡(1,2) . Recall that 𝑓𝑡(𝑛,𝑚) is the log forward rate for a risk-free investment
between 𝑡 + 𝑛 and 𝑡 + 𝑛 + 𝑚. Express the forward rates in annualized percentage rates.
Suppose the expectations hypothesis holds. In which years was the market expecting
rates to decline versus increase? Are the expectations correct?

3. Stock-Bond Correlation. Compute the daily bond returns using data from the previous
question. For each maturity 𝑛 (years), approximate the daily return by the daily changes in
(𝑛)
𝑒 −𝑛𝑦𝑡
SVENYn: 𝑅𝑡(𝑛) = (𝑛) , where 𝑅𝑡(𝑛) is the daily return on the 𝑛-year bond, 𝑛 denotes years
−𝑛𝑦
𝑒 𝑡 −1
to maturity, 𝑡 denotes the date of observation, 𝑡 − 1 denotes the previous day, and 𝑦 𝑡(𝑛) is
SVENYn at date 𝑡. If the previous calendar day has no data, treat 𝑡 − 1 as the previous day
that has data. Note that SVENYn is measured in percentage points and you need to divide
it by 100 before plugging it into the formula. Daily S&P 500 returns are available on the
course website. Use “vwretd” for the exercise. For each month, compute the correlation
between the S&P 500 and 20-year Treasury bond daily returns within the month. Plot the
time series of monthly stock-bond return correlations. Do you find any interesting patterns?

2 Bonus
Consider[a consumption-based
] asset pricing model. A representative investor maximizes lifetime
∑∞ −𝜌𝑡
utility E 𝑡=0 𝑒 ln𝐶𝑡 . The consumption-based asset pricing theory states that the stochastic
discount factor is 𝑀𝑡 = 𝑒 −𝛿𝑡 𝐶1𝑡 , and cashflows at 𝑡 + 𝑇 are discounted by 𝑀𝑀𝑡+𝑇
𝑡
.

1. Assume that the growth of consumption is i.i.d normal:

ln𝐶𝑡+1 = 𝜇 𝑐 + ln𝐶𝑡 + 𝜎 𝑐 𝜀𝑡+1


𝑐
, 𝑐
𝜀𝑡+1 ∼ N (0, 1); Cov(𝜀𝑡𝑐 , 𝜀 𝑐𝑠 ) = 0, ∀𝑡 ≠ 𝑠. (2)

𝜇 𝑐 and 𝜎 𝑐 are positive parameters.

(a) Compute the 1-period risk-free log yield.

2
2025 Spring Instructor: Zehao Li

(b) A stock pays dividend 𝐷 𝑡 in each period 𝑡. The dividend process is

ln 𝐷 𝑡+1 = 𝜇 𝑑 + ln 𝐷 𝑡 + 𝜎 𝑑 𝜀𝑡+1
𝑑
, 𝑑
𝜀𝑡+1 ∼ N (0, 1); Cov(𝜀𝑡𝑑 , 𝜀 𝑠𝑑 ) = 0, ∀𝑡 ≠ 𝑠. (3)

Furthermore, Corr(𝜀𝑡𝑐 , 𝜀𝑡𝑑 ) = 𝜌 ∈ [−1, 1] and Corr(𝜀𝑡𝑐 , 𝜀 𝑠𝑑 ) = 0, ∀𝑡 ≠ 𝑠. Assume that


( )2
𝑐 (𝜎 𝑐 ) 2 + 𝜎 𝑑
𝑑
𝛿+𝜇 −𝜇 − + 𝜎 𝑐 𝜎 𝑑 𝜌 > 0.
2
[ ]
𝑓 𝑓
Compute the equity premium, i.e., E𝑡 ln 𝑃𝑡+1𝑃+𝐷
𝑡
𝑡+1
− 𝑟 𝑡 , where 𝑟 𝑡 is the 1-period risk-
free log yield. How does the equity premium depend on 𝜌? What’s the intuition for
the relationship?

You might also like