12.07.2015 Views

"Frontmatter". In: Analysis of Financial Time Series

"Frontmatter". In: Analysis of Financial Time Series

"Frontmatter". In: Analysis of Financial Time Series

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

354 VECTOR TIME SERIES3. Σ 12 = 0 if and only if x 1 and x 2 are independent.4. The random variable y = (x − µ) ′ Σ −1 (x − µ) follows a chi-squared distributionwith m degrees <strong>of</strong> freedom.5. The conditional distribution <strong>of</strong> x 1 given x 2 = b is also normally distributed as(x 1 | x 2 = b) ∼ N p [µ 1 + Σ 12 Σ −122 (b − µ 2), Σ 11 − Σ 12 Σ −122 Σ 21].The last property is useful in many scientific areas. For instance, it forms the basisfor time series forecasting under the normality assumption and for recursive leastsquares estimation.EXERCISES1. Consider the monthly log stock returns, in percentages and including dividends,<strong>of</strong> Merck & Company, Johnson & Johnson, General Electric, General Motors,Ford Motor Company, and value-weighted index from January 1960 to December1999; see the file “m-mrk2vw.dat,” which has six columns in the order listedbefore.(a) Compute the sample mean, covariance matrix, and correlation matrix <strong>of</strong> thedata.(b) Test the hypothesis H 0 : ρ 1 = ··· = ρ 6 = 0, whereρ i is the lag-i crosscorrelationmatrix <strong>of</strong> the data. Draw conclusion based on the 5% significancelevel.(c) Is there any lead-lag relationship among the six return series?(d) Perform a principal component analysis <strong>of</strong> the data using the sample covariancematrix.(e) Perform a principal component analysis <strong>of</strong> the data using the sample correlationmatrix.(f) Perform a factor analysis on the data. Identify the number <strong>of</strong> common factors.Obtain estimates <strong>of</strong> factor loadings using both the principal component andmaximum likelihood methods.2. The Federal Reserve Bank <strong>of</strong> St Louis publishes selected interest rates and U.S.financial data on its Web site:http://www.stls.frb.org/fred/index.htmlConsider the monthly 1-year and 10-year Treasury constant maturity rates fromApril 1953 to October 2000 for 571 observations; see the file “m-gs1n10.dat.”The rates are in percentages.(a) Let c t = r t − r t−1 be the change series <strong>of</strong> the monthly interest rate r t . Builda bivariate autoregressive model for the two change series. Discuss the implications<strong>of</strong> the model. Transform the model into a structural form.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!