Thesis (Ph. D.)--University of Washington, 2000. / Vita. Includes bibliographical references (leaves 88-93).

This dissertation is composed of three essays which examine different topics in empirical asset pricing. Chapter 1 is the result of joint work with Andrew Ang and William Goetzmann. First, we document that American university and college endowments have shifted their asset allocations from stocks to alternative investments. By the end of the sample, the average endowment holds close to one third of its portfolios in private equity and hedge funds. What are the expectations of future returns that can explain these changes in portfolio holdings? Fitting a simple asset allocation model using Bayesian methods, we estimate that at the end of 2012, the average university expects its private equity investments to outperform a portfolio of conventional assets by 3.9% per year and hedge funds to outperform by 0.7% per year. These out-performance beliefs have increased over time, reaching their peak at the end of our sample. There is also significant cross-sectional heterogeneity in our results. Private institutions, universities with large endowments and high spending rates, and those that rely more on their asset holdings to meet operational budgets tend to expect higher alphas from alternative investments. Chapter 2 examines to what extent commodity prices have contributed to the inflation volatility experienced by the Chilean economy in recent years. First, I show that oil is the commodity that is most correlated with future inflation and inflationary expectations. Next, I use a Gaussian affine term structure model with observable macroeconomic factors to quantitatively study how shocks to oil prices affect bond yields and inflation expectations. I find a statistically significant but economically modest effect. An increase in the price of oil of 20% raises one-year inflation expectations by 25 basis points, while five-year expectations increase only by 8 basis points. The results suggest that central banks could benefit from paying attention to commodity prices when setting monetary policy. Finally, Chapter 3 studies both theoretically and empirically whether market expectations on the health of the financial sector affect stock returns. Prior literature shows that the ratio of intermediary equity to GDP predicts future market returns and is a priced risk factor in the cross-section of stock returns. Here, I extend this work and show that expectations of large declines in the capital of financial institutions can also help explain equity returns. Specifically, I show that different measures of intermediary equity tail-risk are priced in the cross-section. Firms that load on this financial tail-risk factor have lower expected returns. Motivated by these facts, I develop an intermediary asset pricing model where the financial sector's net worth is subject to large negative exogenous shocks. I calibrate the model to U.S. data and find that stocks that do well when disaster risk is high earn significantly lower returns, thus providing theoretical support to my findings. In addition, the model is able to match key asset pricing moments like the equity premium and the volatility of stock returns.

A central topic in empirical asset pricing is how to explain anomalies in various trading horizons. This dissertation contains two essays that study several anomalies in medium-term/long-term investment in the equity market and in high-frequency trading in the foreign exchange market. In the first essay, I propose an investor underreaction model with heterogeneous truncations across time and stocks. In this setting, investors are more attracted to dramatic changes in stock prices than to gradual changes. Continuous information causes signals to be truncated which delays their incorporation into stock prices thus generating momentum. Under the assumption that investors are more attracted to winner stocks and ignore more information in loser stocks, I show that a loser portfolio exhibits stronger momentum and higher profitability than a winner portfolio with the same discreteness level. A trading strategy based on this model yields high alphas and Sharpe ratios. Evidence from social media trends aligns well with this model. In the second essay, I develop multivariate logistic models to explain the short-term offer price movement of the currency pair EUR/USD from the EBS limit order book. Using logistic regression based methods, I study the impact of various market microstructure factors on offer price changes in the next second. The empirical results show explanatory power for the testing sample up to 45% and a true positive rate of the prediction up to 87%. The model reveals interesting mechanisms for the underlying driving forces of the tick-by-tick currency price movement.

Mode of access: World Wide Web. / Thesis (Ph. D.)--University of Hawaii at Manoa, 2004. / Includes bibliographical references (leaves 149-151). / Electronic reproduction. / Also available by subscription via World Wide Web / xi, 158 leaves, bound ill. 29 cm

The Market Timing - Buy and Hold (MT-BH) is introduced, tested against widely accepted performance models of market timing and tested if implamentation is possible. The MT-BH metric measures the condition of engaging in market timing strategies relative to buy and hold investing across an equity market. The metric provides an alternative explanation to why market timing results of investors and managers vary through time and across different equity markets. This dissertation examines how the is correlated with traditional market timing measures of the Treynor and Sharpe ratios over the 1995-2010 time period and how it affects widely used measures of regression based market timing models of Treynor- Mazuy and Henriksson-Merton. The Market Timing - Buy and Hold (MT-BH) metric can be applied to any equity market over any time period to condition the market timing skill of money managers in any equity market around the world. The final accomplishment of this dissertation is to determine if readily available finance and macro-economic variables can help investors determine which years are more favorable to pursue market timing strategies and which years favor buy and hold investing. When real GDP growth rates, inflation rates and PE ratios were low or negative and when dividend yields were high, market timing strategies were favorable across 44 country market indexes from 1994-2008. These results were robust to country level of development, negative market return years and other control variables. The conditions for pursing market timing strategies were time variant and detectable with macro-economic and finance variables. The MT-BH metric allows investors and brokers to determine when to switch from buy and hold investing to a market timing strategy using macro-economic and financial variables and helps to explain why market timing skill of managers is rarely found to be persistent. / by William Fount Johnson, III. / Thesis (Ph.D.)--Florida Atlantic University, 2011. / Includes bibliography. / Electronic reproduction. Boca Raton, Fla., 2011. Mode of access: World Wide Web.

This dissertation aims to investigate the asset pricing implications of the stock option's implied volatility term structure. We mainly focus on two directions: the volatility term structure of the market and the volatility term structure of individual stocks. The market volatility term structure, which is calculated from prices of index options with different expirations, reflects the market's expectation of future volatility of different horizons. So the market volatility term structure incorporates information that is not captured by the market volatility itself. In particular, the slope of the volatility term structure captures the expected volatility trend. In the first part of the thesis, we investigate whether the market volatility term structure slope is a priced source of risk or not. We find that stocks with high sensitivities to the proxies of the VIX term structure slope exhibit high returns on average. We further estimate the premium for bearing the VIX slope risk to be approximately 2.5% annually and statistically significant. The effect cannot be explained by other common risk factors, such as the market excess return, size, book-to-market, momentum, liquidity and market volatility. We extensively investigate the robustness of our empirical results and find that the effect of the VIX term structure risk is robust. Within the context of ICAPM, the positive price of VIX term structure risk indicates that it is a state variable which positively affects the future investment opportunity set. In the second part of the thesis, we provide a stylized model that explains our empirical results. We build a regime-switching rare disaster model that allows disasters to have short and long durations. Our model indicates that a downward sloping VIX term structure corresponds to a potential long disaster and an upward sloping VIX term structure corresponds to a potential short disaster. It further implies that stocks with high sensitivities to the VIX slope have high loadings on the disaster duration risk, thus earn higher risk premium. These implications are consistent with our empirical results. In the last part, we study the relationship between individual stock's volatility term structure and the stock's future return. We use a measure of stock's implied volatility term structure slope, defined as the difference between 3-month and 1-month implied volatility from at-the-money options, to demonstrate that option prices contain important information for the underlying equities. We show that option volatility term structure slopes are significant in explaining future equity returns in the cross-section. And we further find evidence that the implied volatility term structure is a measure of event risk: firms with the most negative volatility term structure are those for which the market anticipates news that may affect stock price within one month. Relevant events include, but are not limited to, earnings announcements.

The first essay examines the joint determination of the contract for a private equity (PE) fund manager and the equilibrium risk premium of the PE fund. My model relies on two realistic features of PE funds. First, I model agency frictions between PE fund's investors and manager. Second, I model the illiquidity of PE fund investments. To alleviate agency frictions, compensation to the manager becomes sensitive to the PE fund performance, which makes investors excessively hold the PE fund to hedge the manager's fees. This induces a negative effect on the risk premium in equilibrium. For the second feature, I add search frictions in the secondary market for PE fund's shares. PE fund returns also contain a positive illiquidity premium since investors internalize the possibility of holding sub-optimal positions in the PE fund. Thus, my model delivers a plausible explanation for the inconclusive findings of the empirical literature regarding PE funds' performance. Agency conflicts deliver a lower risk-adjusted performance of PE funds, while illiquidity risk can raise it. In the second essay, coauthored with Andrew Ang and Pierre Collin-Dufresne, we investigate how often investors should adjust asset class allocation targets when returns are predictable and updating allocation targets is costly. We compute optimal tactical asset allocation (TAA) policies over equities and bonds. By varying how often the weights are reset, we estimate the utility costs of different frequencies of TAA decisions relative to the continuous optimal Merton (1971) policy. We find that the utility cost of infrequent switching is minimized when the investor updates the target portfolio weights annually. Tactical tilts taking advantage of predictable stock returns generate approximately twice as much value as those market-timing bond returns. In the third essay, also coauthored with Andrew Ang and Pierre Collin-Dufresne, we revisit the question of a pension sponsor's optimal asset allocation in the presence of a downside constraint and the possibility for the pension sponsor to contribute money to the pension plan. We analyze the joint problem of optimal investing and contribution decisions, when there is disutility associated with contributions. Interestingly, we find that the optimal portfolio decision often looks like a ``risky gambling" strategy where the pension sponsor increases the pension plan's allocation to risky assets in bad states. This is very different from the traditional prediction, where in economy downturns the pension sponsor should fully switch to the risk-free portfolio. Our solution method involves a separation of the pension sponsor's problem into a utility maximization problem and a disutility minimization one.

How are the prices of financial assets determined? In this dissertation, I test various theories empirically, focusing on several classes of bonds. In the first chapter, I test whether asset prices reflect the risk-exposures of financial intermediaries in a setting that is well suited to tackling concerns about omitted risk factors. I analyze catastrophe bonds whose cash flows are linked to the occurrence of natural disasters and find that 71% of the variation in their expected returns can be explained by a theoretically-motivated measure of financial intermediaries’ marginal rate of substitution. Assuming that natural disasters are independent of aggregate wealth, this pricing result is inconsistent with any explanation based on macroeconomic risk factors. However, the result is consistent with intermediary asset pricing models that suggest that financial intermediaries are marginal investors in capital markets. I also show that the premium on natural disaster risk has decreased significantly in recent years and has become less responsive to the occurrence of disasters, suggesting that intermediaries’ access to outside capital has improved over time. In the second chapter, which is coauthored with Robert J. Hodrick, we examine the statistical term structure model of Cochrane and Piazzesi (2005) and its affine counterpart, developed in Cochrane and Piazzesi (2008), in several out-of-sample analyzes. The model’s one-factor forecasting structure across bonds with two, three, four, and five years to maturity characterizes the term structures of additional major currencies in samples ending in 2003. In post-2003 data such one-factor structures again characterize each currency’s term structure, but we reject equality of the coefficients across the two samples. We derive currency return forecasting implications from the Cochrane and Piazzesi (2008) affine model showing that the term structure forecasting variables in each currency should predict cross-currency investments, but we find no support for these predictions in either pre-2004 or post-2003 data, whereas the interest differentials do predict currency returns. Here too, though, we find strong evidence of parameter instability as the parameter estimates on the interest differentials change sign. In recursive out-of-sample forecasts of excess rates of return on bonds in each currency, the Cochrane and Piazzesi (2008) term structure forecasting models fail to beat forecasts from the historical average excess rates of return. Graphical analysis indicates that the instability in the forecasting models’ parameters begins in the global financial crisis.

Mehra and Prescott's (1985) equity premium puzzle has stirred continued debate on just why the average return on equity has been so high relative to the risk-free rate. Recent work by Backus, Gregory, and Zin (1989), Knez and Snow (1992), and Trainor (1992) have also documented a liquidity premium puzzle. In addition, Fama and French (1992) have found that beta has no explanatory power in explaining an asset's excess return. These studies point out that current financial models are unable to explain even the most basic premise that assets with greater risk have higher returns. The question that now arises is why are these financial models failing to explain excess returns? One obvious answer to this question which has been completely ignored is that the proxy being used to define risk is wrong. It is the contention of this proposal that investors are concerned about buying into an asset and subsequently experiencing a sequence of below average or negative returns. Under this premise, using the variance of returns as a measure of risk is inadequate and a new risk measure must be derived. This study demonstrates that measuring the deviation of an investor's wealth level from buying a risky asset in relation to what an investor's wealth level would have been from buying a risk-free asset discerns both the deviation of returns and the propensity of returns to sequence. It is then shown that sequencing risk and the slope of the term structure are integrally related. Specifically, the steeper the yield curve, the greater sequencing risk will be priced since a negative sequence could result in forced borrowing by investors when rates are high to maintain a constant consumption rate. Empirically, it is shown that measuring an asset's risk by the contribution it makes to a portfolio's propensity to sequence rather than to a portfolio's variance more accurately explains portfolio returns within a CAPM type framework. Additionally, size does not usurp the explanatory power of this new beta. Surprisingly, it is found that the explanatory power of the traditional beta and size are contingent upon the slope of the term structure being fairly flat. The wealth beta seems to be unaffected. The conclusion of the study suggests that current financial models are seriously flawed due to the erroneous definition and mis-measurement of risk. / Ph. D.

[Truncated abstract] This thesis examines competing asset-pricing models in Australia with the goal of establishing the model which best explains cross-sectional stock returns. The research employs Australian equity data over the period 1980-2001, with the major analyses covering the more recent period 1990-2001. The study first documents that existing asset-pricing models namely the capital asset pricing model (CAPM) and domestic Fama-French three-factor model fail to meet the widely applied Merton?s zero-intercept criterion for a well-specified pricing model. This study instead documents that the US three-factor model provides the best description of Australian stock returns. The three US Fama-French factors are statistically significant for the majority of portfolios consisting of large stocks. However, no significant coefficients are found for portfolios in the smallest size quintile. This result initially suggests that the largest firms in the Australian market are globally integrated with the US market while the smallest firms are not. Therefore, the evidence at this point implies domestic segmentation in the Australian market. This is an unsatisfying outcome, considering that the goal of this research is to establish the pricing model that best describes portfolio returns. Given pervasive evidence that liquidity is strongly related to stock returns, the second part of the major analyses derives and incorporates this potentially priced factor to the specified pricing models ... This study also introduces a methodology for individual security analysis, which implements the portfolio analysis, in this part of analyses. The technique makes use of visual impressions conveyed by the histogram plots of coefficients' p-values. A statistically significant coefficient will have its p-values concentrated at below a 5% level of significance; a histogram of p-values will not have a uniform distribution ... The final stage of this study employs daily return data as an examination of what is indeed the best pricing model as well as to provide a robustness check on monthly return results. The daily result indicates that all three US Fama-French factors, namely the US market, size and book-to-market factors as well as LIQT are statistically significant, while the Australian three-factor model only exhibits one significant market factor. This study has discovered that it is in fact the US three-factor model with LIQT and not the domestic model, which qualifies for the criterion of a well-specified asset-pricing model and that it best describes Australian stock returns.