- 學位論文
波動性與熊市：以投資組合為例
Volatility and Bear Market: The Case of US Portfolio
摘要
股票市場是反映經濟發展狀況的重要指標之一，而股票市場的波動性也是很多投資者關注的焦點。近年來，各種全球事件對股市產生重大影響，其中包括2019年開始的中美貿易戰，2020年開始的Covid-19大爆發以及2022年開始的烏俄戰爭。這些事情導致股市進入長時間的股市狀態，其波動性引起投資者更關注這個議題。因此本研究以美國投資組合為例，探討市場波動性是否會引發投資組合在未來進入熊市。本研究期間為1971年3月至2022年7月，使用的波動性由Nasdaq指數以及S&P500指數以realized volatility衡量股市波動性，并採用Kenneth R. French Data library 所提供的根據規模和賬面市值形成的6個投資組合以Pagan & Sossounov（2003）無母數方法建構熊市序列，最後以Probit模型對進行估計與預測。 研究結果顯示，大盤波動性與熊市發生機率之間關係之間為正向顯著關係。其中Big Value投資組合的預測力最强，預測力達5期；而Small Value投資組合的預測力最弱，只有一期顯著。本研究認爲這是因爲市值小的投資在股市行情不好的時候可以快速抽離資金，但市值大的投資組合若快速抽離資金將導致股市的再次下跌，也會引發投資人的恐慌情緒，因此市值大的投資組合只能忍受股市的崩跌。 上述結果顯示當大盤波動性產生明顯變動時，股市會有很高機率進入熊市。這一發現表明，波動性加劇是股市潛在下滑的預警信號。因此視爲市場波動性可作爲投資組合的領先指標。此外，爲了檢測大盤波動性的強韌性，本研究加入8個總體經濟變數以及使用子樣本分析來檢視大盤波動性對熊市的預測力是否穩定。結果顯示，加入總經變數後的模型以及子樣本分析結果與全樣本的結果不相上下。 這些發現對投資者和風險管理者具有重要意義。本研究結果提供了一種基於市場波動性的指標，可以幫助投資者在市場不穩定時做出更明智的投資決策，保護資金的同時快樂交易。通過深入研究市場波動性對未來熊市的預測，本研究填補了現有文獻中的知識空白，同時提供了實際應用上的啟示。本研究結果為金融研究和實踐領域提供了新的觀點和策略，同時為未來相關研究提供了新的方向。
並列摘要
The stock market is one of the key indicators reflecting the state of economic development, and the volatility of the stock market is a major concern for many investors. In recent years, various global events have significantly impacted the stock market, including the US-China trade war that started in 2019, COVID-19 pandemic started in 2020 and the Russia-Ukraine conflict started in 2022. These events have contributed to prolonged periods of stock market downturn and increased investor attention to volatility. Therefore, this study examines whether market volatility will trigger a bear market in the future, using a US portfolio as an example. The study period spans from March 1971 to July 2022, and volatility is measured using the Nasdaq and S&P 500 indices with realized volatility as a measure of stock market volatility. Additionally, six investment portfolios are constructed based on size and book-to-market ratio, as provided by the Kenneth R. French Data Library, to form bear market sequences using the nonparametric method proposed by Pagan & Sossounov (2003). Finally, the Probit model is employed for estimation and prediction. The results indicate a significant positive relationship between overall market volatility and the probability of bear market occurrences. Among the portfolios, the Big Value portfolio exhibits the strongest predictive power, with a forecasting horizon of up to five periods, while the Small Value portfolio shows the weakest predictive power, with only one period showing significance. It is believed that this is due to the fact that small value portfolios can withdraw funds quickly when the stock market is not doing well, but large value portfolios can only endure a collapse in the stock market if the rapid withdrawal of funds will lead to another fall in the stock market, which will also trigger panic among investors. The above results demonstrate that when overall market volatility experiences significant changes, there is a high probability of the stock market entering a bear market. This finding suggests that increased volatility is an early warning sign of a potential downturn in the stock market. Market volatility is therefore considered to be a leading indicator for investment portfolios. Additionally, to assess the robustness of overall market volatility, this study incorporates eight macroeconomic variables and conducts sub-sample analysis to examine whether the predictive power of overall market volatility on bear markets remains stable. The results indicate that the model incorporating macroeconomic variables and the sub-sample analysis yield comparable results to those obtained from the full sample. These findings have important implications for investors and risk managers. The results of this study provide an indicator based on market volatility that can help investors make more informed investment decisions in times of market instability, protect their money and trade happily. By conducting in-depth research on the predictive nature of market volatility on future bear markets, this study fills a knowledge gap in the existing literature while offering practical insights for real-world applications. The results of this study provide new perspectives and strategies for the field of financial research and practice, as well as new directions for future related research.