Analysis of Stock Bubbles: Detection, Impact on Market Stability, and Portfolios’ Reactions

Details

Supervisors

Candelon, Bertrand

Faculty

Louvain School of Management

Degree label

Master [120] : ingénieur de gestion, à finalité spécialisée

Abstract

Stock bubbles have been the triggering event of many crises. Therefore, it is crucial to understand these events and prevent them. The objectives of this thesis are to learn how to identify these phenomena, understand their consequences on market stability, and investigate their effect on investment portfolios. This paper is guided by the following research question: “Which strategies help in discerning stock bubbles, and to what extent do these phenomena impact overall market stability and the performance of investment portfolios?”. For this purpose, various methods were used. Firstly, regarding the first objective, we used various bubbles detection algorithms on R such as the GSADF and structural break tests. We also looked at deviations from historical trends and fundamental valuations, and differences in return distribution. Using the dot-com bubble as a practical case study and specific companies like Microsoft, we gained insight into how to cross results from different tests to identify bubbles. In addition, we also performed tests to assess the duration and intensity of the dot-com bubble on Microsoft and looked at factors that could have influenced the bubble. We gained valuable insights showing a link between the NASDAQ index around the dot-com bubble and the USA’s GDP, the consumer opinion index, and the consumer price index. Then, regarding the second objective, through GARCH analysis, we observed a positive correlation between the volatility of Microsoft, the NASDAQ and the S&P 500 around the bubble. We also gained insight into the positive correlations that exist between different asset classes, emphasizing the potential propagation of these events. Finally, regarding the third objective, we constructed four different portfolio types: an aggressive one, a conservative one, a balanced one, and an international diversified one. We then backtested them between 1995 and 2005 and observed that the balanced one had the best risk-return, in line with the benefits of diversification seen in the literature review. We compared this portfolio with a risk-free rate through the use of the Sharpe ratio, and found that the balanced portfolio was offering a better risk-return tradeoff.