Fractal image compression is based on the representation of an image by contractive transforms whose xed points are close to the original image. In the encoding of fractal image compression, most of the time is spent on nding the close match between a range block and a large pool of domain blocks. In this thesis, we use the intra-block variances distributions of domain blocks to reduce the searching space. For nding a close match we need only search the domain blocks whose maximal intra-block variance quadrants are the same as that of the range block in nding a close match. The experiment results show that our algorithm can reduce much encoding time with only slight loss of quality.

Real Estate Investment Trusts is suggested as an attractive addition to mixed-asset portfolio. This study develops several hypothesized portfolio and tests whether REITs can actually increase diversification benefits of investors. We use mean-variance spanning test by Kan and Zhou (2008) to examine whether adding a REITs into portfolio can significantly expand efficient frontier in either global minimum variance portfolio or tangency portfolio. We assume our investors hold portfolio in the four markets, namely Japan, Singapore, Taiwan and US markets for period from March 2005 to February 2008. Three hypotheses are tested under various assumed conditions. The first hypothesis, which is REITs can provide diversification benefit, is confirmed in all these four markets. In addition, we find, for Taiwan domestic investors, holding international REITs in their portfolio rather than only Taiwan¡¦s REITs will provide more diversification benefit. The second hypothesis, which is holding period will affect diversification benefit, is not supported. However, this could be resulted from a test of short period in this study. The final hypothesis, which is different investment portfolio will affect the diversification benefit of RETIs for Taiwan domestic investors, is confirmed. Our results also suggest that expanding of efficient frontier are mainly from global minimum variance portfolio rather than tangency portfolio.

Thesis--University of Oklahoma. / Includes bibliographical references (leaves 41-44).

Thesis (Ph. D.)--University of Texas at Austin, 2000. / Vita. Includes bibliographical references (leaves 316-324). Available also in a digital version from Dissertation Abstracts.

Thesis (Ph. D.)--University of Chicago, Department of Statistics, June 2001. / Includes bibliographical references. Also available on the Internet.

Thesis (M. Phil.)--University of Hong Kong, 1988.

published_or_final_version / Statistics / Master / Master of Philosophy

published_or_final_version / Statistics and Actuarial Science / Master / Master of Philosophy

In contrast to the studies of constant or time-varying correlations between stock and bond returns, in this thesis, I explore the regime-dependent correlations between stock and bond returns. Specifically, I start with a comprehensive asset pricing model, i.e., a regime-switching multifactor model, and then investigate the regime-dependent correlations between stock and bond returns. Based on the BIC, the number of regimes in the regime-switching model is optimally determined to be two. For the two regimes, the directions of the regime-dependent correlations appear to be significantly different. Also, the magnitudes of the regime-dependent correlations are substantially larger in these two regimes than the correlation in the single regime. With my findings in the regime-dependent correlations, I then examine the performance of portfolio strategies. Throughout the in-sample and out-of-sample tests, I find that the two portfolio strategies, regime inferred portfolio and probability implied portfolio, can outperform the benchmark, S&P 500.

No description available.