Scribd
Upload
31 views

Assignment

The document provides instructions for a homework assignment analyzing gold price data from 2003 to 2006. The assignment asks the student to: 1) Plot the gold price time series and test for stationarity and appropriate ARIMA model identification. The results show the gold price series is non-stationary but becomes stationary after first differencing, indicating an ARIMA(p,1,q) model. 2) Estimate the best fitting ARIMA model and use it to forecast gold price changes over the next 6 periods. 3) Estimate an AR(p,d,q)-GARCH(1,1) model to analyze time-varying volatility, and report the parameter estimates and significance from the volatility

Uploaded by

Nitesh Agrawal
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
31 views

Assignment

The document provides instructions for a homework assignment analyzing gold price data from 2003 to 2006. The assignment asks the student to: 1) Plot the gold price time series and test for stationarity and appropriate ARIMA model identification. The results show the gold price series is non-stationary but becomes stationary after first differencing, indicating an ARIMA(p,1,q) model. 2) Estimate the best fitting ARIMA model and use it to forecast gold price changes over the next 6 periods. 3) Estimate an AR(p,d,q)-GARCH(1,1) model to analyze time-varying volatility, and report the parameter estimates and significance from the volatility

Uploaded by

Nitesh Agrawal
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 2

Assignment 1

Imagine you are a gold trader and specifically, you trade gold options. As such you are very aware that volatility is
critical to your understanding of option pricing. Your immediate supervisor asks you some questions concerning
volatility in the market. Your supervisor has provided you with a time series of gold prices and the volume of gold
trading. The data runs from Jan 2003 to May 2006 and can be found in Gold.xls in Assignment 1 folder.
He has asked you to estimate volatilities for gold. As a good student of Fina6282, you first proceed with identifying
the appropriate time series process for the gold futures prices.
Question 1
1A) Provide a time series plot (chart) of the gold futures price.
Discuss
1B) Test the gold futures price time series for
a) Stationarity
b) Plot ACF and PACF functions

Discuss the results of these tests and any implications for your modeling process

1C) First-difference the gold prices and plot again.


Compare to the original undifferenced plots.
1D) Test the differenced gold price time series for
c) Stationarity
d) Plot ACF and PACF

Discuss the results of these tests and any implications for your modeling process.

1D) Identify the appropriate time series model for the gold prices;
a) Determine the appropriate p, d, q.
b) Estimate the ARIMA model(s) you determined above.
c) Test residuals for white noise.
d) In case you have multiple candidate models, pick the best one from the well behaved models. Explain
your reasoning why you think this is the right way to model Gold price series.
e) Write out the final model (with coefficients). Interpret the results.
1E) Forecast changes in gold futures prices 6 periods ahead using your final ARIMA model.
Question 2
2A) Estimate an AR(p,d,q) GARCH(1,1) model depending on the p, d, q you determined above. Provide the
output from the model.
2B) From the AR(p,d,q) GARCH(1,1) output answer the following: What is the GARCH1 parameter estimate and
ARCH1 parameter estimate in the GARCH volatility equation? Are they statistically significant (what p value do
each of them have)?
SUPPORT YOUR ANSWERS WITH APPROPRIATE EXCERPTS FROM SAS OUTPUT.

Attach your SAS code with the answers. Append it to the end of file and save as pdf. Submit via Blackboard.

You might also like