Lu Lan. / "June 2004." / Thesis (Ph.D.)--Chinese University of Hong Kong, 2004. / Includes bibliographical references (p. 50-53). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

In this study, we address two research questions: 1) Can we identify bear market episodes in Australia in the past 20 years? 2) How do investors’ moods change as stock market conditions enter into a bear phase. To address the first question, we use a pattern recognition algorithm, called the penalised LSE approach. By defining bear markets as those stock market regimes where the average returns are statistically significantly negative or below the risk free rate, we are able to detect two bear market periods in Australia in the past 20 years. These are the November 1987 to February 1988 and the April 2000 to May 2000 periods. To address the second question, we study the change in investors’ attitudes to varieties of systematic risk and the aggregate number and dollar value of shares traded in portfolios as a result of the regime switch from pre-bear to bear period. Out of the 7 categories of risk considered in this study, the transition from pre-bear to bear regime in both sample periods had a significant impact mainly on investors’ attitude toward the size risk factor. Investors systematically became more sensitive to firm size as stock market conditions entered into the 1987⁄1988 bear market. In the later sample period, investors’ reaction to firm size was more selective as it depended on the characteristics of the stocks that made up their portfolios. We also find that the regime switches resulted in lower portfolio trading volumes. Based on these results we infer that the November 1987-February 1988 bear market evoked a general sad mood, while the April 2000-May 2000 bear market stirred up both angry and sad feelings in market participants depending on the composition of stocks in their portfolios.