鍾經樊Chung, Ching-Fan臺灣大學:財務金融學研究所葉宗穎Yeh, Chung-YingChung-YingYeh2010-05-112018-07-092010-05-112018-07-092009U0001-1601200917183000http://ntur.lib.ntu.edu.tw//handle/246246/182706摘要一章提出了一個一般化的隨機動態風險模型(stochastic volatility model)，我們稱為「隨機動態風險反饋模型(stochastic volatility feedback model)」，來研究槓桿效果(leverage effect) 與風險反饋效果(volatility feedbackffect)在實證上與經濟上的重要性。在實證方法上，我們使用Gallant and Tauchen (1996)所提出的Efficient Method of Moment(EMM)來估計。在評價實證上的重要性， 我們是比較哪種效果是決定重製風險與報酬cross-correlations 的首要因素。在經濟上的重要性，我們是以動態資產配置的觀點來衡量。主要結果為：風險反饋效果不論在實證上與經濟上的重要性都高於槓桿效果。二章我們研究投資人的動態資產配置問題，其中該投資人的偏好包含了損失趨避(loss aversion)與框架效果(narrow framing)。而股票價格的動態是帶有可預期機制的(predictability)，既為上一章所提的風險反饋效果(volatility feedback effect)。從理論上我們發現損失趨避與框架效果對投資人資產配置的影響是不同的，其中損失趨避會減少投資人對股票的需求，然而框架效果可以放大或減少對股票的需求，而這決定於投資的輸贏的賠率與心理主觀賠率的相對大。主要結果：當有考慮風險反饋效果的可預期機制時，損失趨避與框架效果並不像Barberis, Huang and Santos (2001), Barberis and Thaler (2003)所言的可以解釋equity premium puzzle/participation puzzle。三章我們同樣研究投資人的動態資產配置問題，其中該投資人的信念受到心理情緒因素所影響。我們試圖比較不同型態投資人的投資表現。為了比較的一致性與合理性，我們在不完全訊息架構下擴充Campbell, Chan and Viceira (2003)的動態資產配置模型，使其能包含投資人的信念，進而分析能投資組合與評價投資表現。我們主要發現：樂觀的投資人的投資表現是最高的，其次才是貝氏投資人；樂觀投資人的投資策略是較積極的，以在牛市市場有最大獲利做為投資法。Abstract hapter 1 proposes a general continuous-time stochastic volatility model, the stochastic volatility feedback (SVF) model, to investigate the empirical and economic importance of the leverage and volatility feedback e ects, both of which are the two main explanations for volatility asymmetry. We empirically estimate our model by the Gallant and Tauchen''s (1996) e cient method of moment (EMM) approach using S&P 500 index returns. To assess the empirical importance of the leverage and volatility feedback e ects, we calculate the simulated cross-correlations between returns and volatility under various speci cations and the make a comparison to its realistic ones. To evaluate the economic importance, we perform the dynamic asset allocation model under various settings for the budget constraint and then compare the economic performances for the corresponding optimal investment strategies. Our ndings are as follows. (1) On comparing the magnitude of the two e ects, the volatility feedback e ect dominates the leverage e ect empirically. The volatility feedback e ect is the key determinant to replicate the sample cross-correlations between returns and volatility. The conventional stochastic volatility models with/without jumps fail to replication; (2) the volatility feedback e ect drives the intertemporal hedging demand, in contrast, the leverage e ect has a minor e ect on it; (3) a longer investment horizon or a higher current volatility enhance the volatility feedback e ect; (4) ignoring the volatility feedback e ect would su er from tremendous economic loss.n chapter 2, we study the dynamic asset allocation problem in a general framework where the representative investor''s preferences include narrow framing/loss aversion parametershile the intertemporal budget constraint contains the empirically-validated volatility feedback e ect. The optimal dynamic allocation of wealth among investment,onsumption, and savings can be solved in an analytically tractable way and is found to be profoundly inenced by the interactions between narrow framing/loss aversion and the volatility feedback e ect. In other words, a more elaborated speci cation of the intertemporal budget constraint including the volatility feedback e ect is indispensable to a complete analysis of narrow framing/loss aversion preferences in a dynamic asset allocation problem. In particular, we nd that introducing narrow framing/loss aversion without the volatility feedback e ect does help reduce investment and transfer the weights to savings, which implies that narrow framing/loss aversion can explain the equity premium puzzle as previous literature has suggested. However, the direction of weight shifts reverses noticeably once the volatility feedback e ect is incorporated. It seems that the preferences of narrow framing/loss aversion alone may not explain the equity premium puzzle in our general framework where the intertemporal budget constraint is more carefully speci ed.n chapter 3, the classical nance theory claimed that irrational investors, who misperceive asset returns, would buy high and sell low, causing them to lose their wealth.he recent researches, such as De Long, Shleifer, Summers and Waldman (1991), and Kogan, Ross, Wang and Wester eld (2006) advocated that irrational investors might formortfolio allocations performing a higher growth rates that outgrow that of the Bayesian investor in market quilibrium. This paper addresses this issue in terms of investor''s assetllocation. We conduct the empirical analysis of the investor''s asset allocation decision considering psychological biases. We specify a regime-switching dynamics of the investment opportunity set and use regime predicting and updating procedures to characterizenvestor sentiment. Our ndings are as follows. (i) In comparison with the Bayesian investor, the optimistic one would like to bet on good states of the economy, more aggressively chasing /shorting the assets with higher rewards in good/bad times. In contrast, the pessimistic investor would take the opposite positions to the optimistic one; (ii) the optimistic investor has the best empirical performance of the portfolio allocation andn turn outperforms the Bayesian one. The other irrational investors underperform the Bayesian one in most cases; (iii) while the predictability in the investment opportunity set is removed, the outperformance of the optimistic investor disappears. It suggests that theene t from the intertemporal hedging induced by the return predictability is a potential explanation to the domination of the optimistic investor in our analysis.Contents How does the Volatility Feedback E ect A ect Asymmetric Volatilitynd Dynamic Asset Allocation 1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Volatility Asymmetries and Empirical Cross-Correlation Patterns . . . . . 4.2.1 The Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.2 The Sample Cross-Correlation Pattern Between Return and Volatility 4.2.3 Two Possible Explanations: the Leverage Hypothesis and the Volatilityeedback Hypothsis . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 The Proposed Asymmetric Volatility Model . . . . . . . . . . . . . . . . . 7.3.1 Two Conventional Stochastic Volatility Models . . . . . . . . . . . . 7.3.2 The Stochastic Volatility Feedback Model . . . . . . . . . . . . . . 8.4 Empirical Performance of SVF Model . . . . . . . . . . . . . . . . . . . . 13.4.1 The Estimation of the Di usion Processes . . . . . . . . . . . . . . 13.4.2 The Estimation Results . . . . . . . . . . . . . . . . . . . . . . . . 15.4.3 The Performance of the SVF Model and the Volatility Feedback E ect 17.5 Dynamic Asset Allocation Analysis . . . . . . . . . . . . . . . . . . . . . . 19.5.1 The Portfolio Allocation Problem . . . . . . . . . . . . . . . . . . . 19.5.2 Stock Demand and the Volatility Feedback E ect . . . . . . . . . . 21.5.3 The Economic Bene ts and The Volatility Feedback E ect . . . . . 24.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26.7 Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28.8 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.8.1 Derivation of Optimal Portfolio Weight and Implementation Finitei erence Method . . . . . . . . . . . . . . . . . . . . . . . . . . . 31.8.2 The Derivation of (2.26) . . . . . . . . . . . . . . . . . . . . . . . . 32 Strategic Asset Allocation under Narrow Framing / Loss Aversion andolatility Feedback 47.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47.2 Asset Allocation under Narrow Framing/Loss Aversion . . . . . . . . . . . 51.3 A Simple Example: the Optimal Consumption, Investment and Savingsnder the Black-Scholes Economy . . . . . . . . . . . . . . . . . . . . . . . 56.4 The Alternative Speci cation of the Intertemporal Budget Constraint . . . 58.4.1 The Discrete-Time Volatility Feedback Model . . . . . . . . . . . . 59.4.2 The Continuous-Time Stochastic Volatility Feedback Model . . . . 61.5 Numerical Analysis of Asset Allocation Under the SVF Model . . . . . . . 62.5.1 The Implication of the Volatility Feedback E ect . . . . . . . . . . 62.5.2 The Myopic Demand for Stock . . . . . . . . . . . . . . . . . . . . 65.5.3 The Value Function and Consumption . . . . . . . . . . . . . . . . 70.5.4 The Intertemporal Hedge Demand for Stock . . . . . . . . . . . . . 74.5.5 Investment, Consumption, and Savings . . . . . . . . . . . . . . . . 76.5.6 Summary Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . 77.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Dynamic Asset Allocation with Distorted Beliefs 95.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95.2 The Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97.2.1 Investment Opportunity Set . . . . . . . . . . . . . . . . . . . . . . 97.2.2 Investor Preferences and Optimality Conditions . . . . . . . . . . . 99.3 Solving the Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100.3.1 The Optimal Filtering Algorithm . . . . . . . . . . . . . . . . . . . 100.3.2 Log-Linearizing the Euler Equations . . . . . . . . . . . . . . . . . 101.3.3 Solving the Optimal Portfolio Weights and Consumption-Wealthatio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102.3.4 The Psychological-Biased Filtering Algorithms . . . . . . . . . . . . 104.4 Empirical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106.4.1 Data Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106.4.2 The Empirical Results of the Investment Opportunity Sets . . . . . 108.5 Empirical Asset Allocations . . . . . . . . . . . . . . . . . . . . . . . . . . 110.5.1 Asset Allocations under the Bayesain Beliefs . . . . . . . . . . . . . 110.5.2 The E ects of the Distorted Beliefs . . . . . . . . . . . . . . . . . . 112.5.3 The E ects of Predictors on Asset Demands . . . . . . . . . . . . . 113.6 Assessing the Asset Allocations under Distorted Beliefs . . . . . . . . . . . 114.6.1 Performance Measures . . . . . . . . . . . . . . . . . . . . . . . . . 114.6.2 Empirical Performances of Asset Allocation under Distorted Beliefs 115.6.3 Implications of the Empirical Performance Results . . . . . . . . . . 116.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117.8 Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119.9 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123.9.1 Derivations of (3.15) and (3.17) . . . . . . . . . . . . . . . . . . . . 123.9.2 Numerical Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . 125application/pdf2737584 bytesapplication/pdfen-US損失趨避,框架效果, 風險反饋,馬可夫轉換, 投資人情緒, 扭曲的信念dynamic asset allocation, narrow framing/loss aversion, volatility feedbackthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/182706/1/ntu-98-D89723006-1.pdf