McMaster University
DeGroote School of Business
MFIN 701
Assignment 3
Due April 8, 5pm.
Hand in a pdf copy of your computer output and a separate write-up of the answers
to the following questions to Avenue, Assignment 3 by 5pm April 8. Each student’s
write-up should be done independently.
1. (50) The dataset is ass3.dat and consists of industrial production (ip), consumer
prices (p), 3 different interest rates (cd,tb,mr) and capacity utilization (cap).
(a) (10) Plot each series and decide based on the plot if the series is I(0) or I(1).
(b) (40) Using the R package tseries test each data series for a unit root with an
ADF test with k = 2 lags. Consider using the log of a series so that a first
difference results in a growth rate. Report your findings. Do they differ from
(a)?
2. (50) Using VIX (vt) found in VIXCLS.csv as a volatility proxy answer the following
questions
(a) (10) Consider a time series plot vt and discuss why modeling log(vt) is preferred.
(b) (10) Test if log(vt) is stationary and make adjustments if necessary for the
remaining questions. Report your findings.
(c) (10) Using the ACF and PACF identify a set of candidate ARMA(p,q) models
for log(vt).
(d) (10) Using the BIC, select the best model from your identified set in (c)
using arima() and report estimates for this model including a Ljung-Box test
for residual autocorrelation. Assuming the model is stationary what is the
implied long-run mean of log(vt). Is it close to the sample mean of log(vt)?
Be sure to check what the arima() command reports.
(e) (10) Report a 1 period out-of-sample forecast for log(vt).