EF4822 Financial Econometrics

Problem Set 2
Due: March 28, 6pm, Canvas online submission
Please submit in groups and write out the names of each groupmate. If R programming is
used, please include the R commands. The data files needed in this problem set are uploaded
into the same Canvas folder as this file.
Note on online submission: Any file format is accepted as long as it is clear to read. PDF
and Word file format are preferred. You could scan, take photos, or type.

1. Suppose that the simple return of a monthly bond index follows the model

rt = 0.02 + at + 0.2at−2 ,

where at is a white noise series with mean zero and standard deviation σa = 0.025. What are
the mean and variance of the return series rt ? Compute the lag-1 and lag-2 autocorrelations
of rt . Assume that a100 = 0.01 and a99 = −0.02. Compute the 1-step-ahead and 2-step-
ahead forecasts of the return series at the forecast origin t = 100. Compute the associated
forecast errors and the standard deviations of the forecast errors.

2. Suppose that the daily log return of a security follows the model

rt = 0.01 + 0.2rt−2 + at ,

where at is a white noise series with mean zero and variance σa2 = 0.02. What are the mean
and variance of the return series rt ? Compute the lag-1 and lag-2 autocorrelations of rt .
Assume that r100 = −0.01, and r99 = 0.02. Compute the 1-step-ahead and 2-step-ahead
forecasts of the return series at the forecast origin t = 100. Compute the associated forecast
errors and the standard deviations of the forecast errors.

3. Consider the monthly simple returns of the Decile 1 (CAP1RET in the data file),
Decile 2 (CAP2RET in the data file), Decile 9 (CAP9RET in the data file), and Decile 10

1
(CAP10RET in the data file) of U.S. stocks based on market capitalization. The data span
is from January 1970 to December 2008. The data file is m-deciles08.txt (uploaded in the
Problem Set folder in Canvas).
(a) For the return series of Decile 2 and Decile 10, test the null hypothesis that the first 12
lags of autocorrelations are zero at the 5% level. Draw your conclusion.
(b) Build an ARMA model for the return series of Decile 2. Check the estimated model. Is
the estimated model a good/adequate model? Write down the estimated model.
(c) Use the estimated ARMA model to produce 1- to 12-step-ahead forecasts of the series
and the associated standard errors of forecasts.

4. Consider the monthly log returns of U.S. CRSP equal-weighted index from January
1962 to December 1999 for 456 observations. The data file is m-ew6299.txt (uploaded in
the Problem Set folder in Canvas).
(a) Build an AR model for the series. Check the estimated model. Is the estimated model
a good/adequate model? Write down the estimated model.
(b) Build a MA model for the series. Check the estimated model. Is the estimated model a
good/adequate model? Write down the estimated model.
(c) Build an ARMA model for the series and check whether the model is the same as the
model in part (a) or part (b).
(d) Compute 1- to 12-step-ahead forecasts of the AR and MA models built in part (a) and
part (b). Are the forecasts in the AR and MA models similar?