Thesis advisor: Zhijie Xiao / In recent years, quantile regression has achieved increasing prominence as a quantitative method of choice in applied econometric research. The methodology focuses on how the quantile of the dependent variable is influenced by the regressors, thus providing the researcher with much information about variations in the relationship between the covariates. In this dissertation, I consider two quantile regression models where the information set may contain quantiles of the regressors. Such frameworks thus capture the dependence between quantiles - the quantile of the dependent variable and the quantile of the regressors - which I call models of quantile dependence. These models are very useful from the applied researcher's perspective as they are able to further uncover complex dependence behavior and can be easily implemented using statistical packages meant for standard quantile regressions. The first chapter considers an application of the quantile dependence model in empirical finance. One of the most important parameter of interest in risk management is the correlation coefficient between stock returns. Knowing how correlation behaves is especially important in bear markets as correlations become unstable and increase quickly so that the benefits of diversification are diminished especially when they are needed most. In this chapter, I argue that it remains a challenge to estimate variations in correlations. In the literature, either a regime-switching model is used, which can only estimate correlation in a finite number of states, or a model based on extreme-value theory is used, which can only estimate correlation between the tails of the returns series. Interpreting the quantile of the stock return as having information about the state of the financial market, this chapter proposes to model the correlation between quantiles of stock returns. For instance, the correlation between the 10th percentiles of stock returns, say the U.S. and the U.K. returns, reflects correlation when the U.S. and U.K. are in the bearish state. One can also model the correlation between the 60th percentile of one series and the 40th percentile of another, which is not possible using existing tools in the literature. For this purpose, I propose a nonlinear quantile regression where the regressor is a conditional quantile itself, so that the left-hand-side variable is a quantile of one stock return and the regressor is a quantile of the other return. The conditional quantile regressor is an unknown object, hence feasible estimation entails replacing it with the fitted counterpart, which then gives rise to problems in inference. In particular, inference in the presence of generated quantile regressors will be invalid when conventional standard errors are used. However, validity is restored when a correction term is introduced into the regression model. In the empirical section, I investigate the dependence between the quantile of U.S. MSCI returns and the quantile of MSCI returns to eight other countries including Canada and major equity markets in Europe and Asia. Using regression models based on the Gaussian and Student-t copula, I construct correlation surfaces that reflect how the correlations between quantiles of these market returns behave. Generally, the correlations tend to rise gradually when the markets are increasingly bearish, as reflected by the fact that the returns are jointly declining. In addition, correlations tend to rise when markets are increasingly bullish, although the magnitude is smaller than the increase associated with bear markets. The second chapter considers an application of the quantile dependence model in empirical macroeconomics examining the money-output relationship. One area in this line of research focuses on the asymmetric effects of monetary policy on output growth. In particular, letting the negative residuals estimated from a money equation represent contractionary monetary policy shocks and the positive residuals represent expansionary shocks, it has been widely established that output growth declines more following a contractionary shock than it increases following an expansionary shock of the same magnitude. However, correctly identifying episodes of contraction and expansion in this manner presupposes that the true monetary innovation has a zero population mean, which is not verifiable. Therefore, I propose interpreting the quantiles of the monetary shocks as having information about the monetary policy stance. For instance, the 10th percentile shock reflects a restrictive stance relative to the 90th percentile shock, and the ranking of shocks is preserved regardless of shifts in the shock's distribution. This idea motivates modeling output growth as a function of the quantiles of monetary shocks. In addition, I consider modeling the quantile of output growth, which will enable policymakers to ascertain whether certain monetary policy objectives, as indexed by quantiles of monetary shocks, will be more effective in particular economic states, as indexed by quantiles of output growth. Therefore, this calls for a unified framework that models the relationship between the quantile of output growth and the quantile of monetary shocks. This framework employs a power series method to estimate quantile dependence. Monte Carlo experiments demonstrate that regressions based on cubic or quartic expansions are able to estimate the quantile dependence relationships well with reasonable bias properties and root-mean-squared errors. Hence, using the cubic and quartic regression models with M1 or M2 money supply growth as monetary instruments, I show that the right tail of the output growth distribution is generally more sensitive to M1 money supply shocks, while both tails of output growth distribution are more sensitive than the center is to M2 money supply shocks, implying that monetary policy is more effective in periods of very low and very high growth rates. In addition, when non-neutral, the influence of monetary policy on output growth is stronger when it is restrictive than expansive, which is consistent with previous findings on the asymmetric effects of monetary policy on output. / Thesis (PhD) — Boston College, 2009. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Economics.

Estudos sobre mobilidade intergeracional de educação sugerem que países subdesenvolvidos apresentam menor mobilidade intergeracional que países desenvolvidos e especificamente para o Brasil, o grau de persistência estimado é ao redor de 0.7, podendo apresentar diferentes graus ao longo da distribuição de educação. Este estudo apresenta uma nova abordagem para a mensuração da mobilidade intergeracional utilizando Regressões Quantílicas. Especificamente, é proposta uma medida de distância entre os quantis condicionais para analisar a mobilidade intergeracional. Como resultado, é obtido um conjunto de matrizes que descrevem o padrão da mobilidade intergeracional em diferentes pontos da distribuição condicional de escolaridade. Utilizando dados para o Brasil, encontra-se que a mobilidade intergeracional tende a ser maior nas caudas da distribuição de escolaridade para filhos e filhas relativo à educação de pais e mães. Comparando filhos e filhas, os filhos tendem a ter menor mobilidade intergeracional que as mulheres relativo à educação de seus pais. Além do mais, a educação das mães tem maior efeito em magnitude do que a educação dos pais tanto para filhos quanto para as filhas. Também se encontrou que a educação dos filhos depende mais da educação do pai e a educação das filhas depende mais da educação das mães, indicando que os filhos tendem a ter educação similar à de seus pais e as filhas tendem a ter educação similar à de suas mães. / Studies on intergenerational educational mobility suggest that underdevelopment countries presents lower intergenerational mobility than developed countries and specifically for Brazil, the estimated degree of persistence is around 0.7 with possible different degrees on the overall distribution of education. This study presents a new approach to measuring intergenerational mobility using quantile regression. Specifically, it is proposed the use of a measure of distance between conditional quantiles to analyze intergenerational mobility. As a result, is obtained a set of matrices which describe the patterns of intergenerational mobility at different points of the conditional distribution of schooling. Using Brazilian Data (PNAD 1996) it is found that intergenerational mobility seems to be higher at the tails of the distribution of schooling for sons and daughters relative to father\'s and mother\'s education. Comparing each other, sons tend to have less mobility than daughters relative to father\'s education. Moreover, mother\'s education has stronger effects than father\'s on both sons and daughters education. It was also found that son\'s education depends more on father\'s education and daughter\'s education depends more on mother\'s education, indicating that sons tends to have education similar to their fathers and daughters tends to have education similar to their mothers.

Zero-inflated outcomes are common in biomedical studies, where the excessive zeros indicate some special but undetectable events. Quantile regression is potentially advantageous in analyzing zero-inflated outcomes due to two reasons. First, compared to parametric models such as the zero-inflated Poisson and two-part model, quantile regression gives robust and accurate estimation by avoiding likelihood specification and can capture the tail events and heterogeneity over the outcome distribution. Second, while the mean-based regression may be misinterpreted for a zero-inflated outcome, the interpretation of quantiles is naturally compatible with the underlying process that such an outcome intends to measure. Unfortunately, uncorrected linear quantile regression is not directly applicable because of two reasons. First, the feasibility of estimation and validity of inference of quantile regression require the conditional distribution of outcomes to be absolutely continuous, which is violated due to zero-inflation. Second, direct quantile regression implicitly assumes a constant chance to observe a positive outcome, but the degree of zero-inflation varies with the covariates in most cases. Thus the conditional quantile function of the outcome depends on the covariates in a nonlinear fashion. To analyze the zero-inflated outcomes by taking advantage of the merits of quantile regression, we propose a novel quantile regression framework that can address all the issues above. In the first part of this dissertation, we propose a two-part model that comprises a logistic regression for the probability of being positive, and a linear quantile regression for the positive part with subject-specific zero-inflation adjusted. Inference on the estimated conditional quantile and covariate effect are not trivial based on such a two-part model. We then develop an algorithm to achieve a consistent estimation of the conditional quantiles, while circumventing the unbounded variance at the quantile level where the conditional quantile changes from zero to positive. Furthermore, we develop an inference tool to determine the quantile treatment effect associated with a covariate at a given quantile level. We evaluate the proposed method and compare it with existing approaches by simulation studies and a real data analysis aimed at studying the risk factors for carotid atherosclerosis. In the second part, based on the proposed two-part model mentioned above, we develop ZIQRank, a zero-inflated quantile rank-score based test to detect the difference in distributions. The proposed test extends the local inference in the first part to a simultaneous one. It is powerful to handle zero-inflation and heterogeneity simultaneously. It comprises a valid test of logistic regression for the zero-inflation and rank-score based tests on multiple quantiles for the positive part with zero-inflation adjusted. The p-values are combined with a procedure selected according to the extent of zero-inflation and heterogeneity of the data. Simulation studies show that compared to existing tests, the proposed test has a higher power in detecting differential distributions. Finally, we apply the ZIQRank test to a human scRNA-seq data to study differentially expressed genes in Neoplastic and Regular cells. It successfully discovers a group of crucial genes associated with glioma, while the other methods fail to do so. In the third part, we extend the proposed two-part quantile regression model for zero-inflated outcomes and the ZIQRank test to analyze longitudinal data. Each part of the proposed two-part model is modified as a marginal longitudinal model (GEE), conditioning on the outcome at the previous time point and its zero/positive status. We apply the model and the test to study the effect of a recommender system aimed at boosting user engagement of a suite of smartphone apps designed for depressed patients. Our novel model framework demonstrates a dominating performance in model fitting, prediction, and critical feature detection, compared to the existing methods.

The objective of this study is to discover the non-linear effect of market sentiment to characteristic factor returns. We run ¡¥Quantile Regression¡¦ to help us extract useful information and design an effective strategy. Based on the quantitative investment method, using the platform of Multi-Factor Model (MFM), we attempt to construct a portfolio and enhance portfolio performance. If the market-sentiment variable increases performance, we could conclude that some characteristic factors in a high sentiment period will perform better or worse in the next period. What is the market or investor sentiment? It is still a problem in the finance field. There is no co-definition or consensus so far. We do our best to collect the indirect data, such as transaction data, price and volume data, and some indicators in other studies, VIX, put/call ratio and so on. Then, we test the proxy variables independently, and obtain some interesting results. The market turnover, the ratio of margin lending on funds/ margin lending on securities, and the growth rate of consumer confidence index have significant effects on some of the characteristic factors. This holds that some market sentiment variables could influence stocks with certain characteristics, and the factor timing approach could improve portfolio performance under examination by information ratio.

This study investigates investment behavior of Taiwan market participants from different aspects of measure, especially with regard to their tendency to forming herding behavior. By applying concepts of Cross-Sectional Absolute Dispersions (CSAD), we find significant evidence of herding behavior in the Taiwan market. Evidences suggest that the herding formation in Taiwan market is strongly influenced by the US market and we should not ignore the impact of globalization. With regard to the issue of financial crises, we find no herding behavior during the 1998 Asian Crisis but partial evidence shows that herding activities may be influenced by crisis during the 2000 Internet Bubble and 2008 Sub-prime Crisis in the Taiwan market. Moreover, all empirical results are reexamined using Quantile analysis to avoid potential bias in estimations. Finally, results from applying herding behavior in portfolio management indicate that investing in stocks of lower liquidity and volatility can reduce the risk of portfolios.

Existing papers have shown that innovation and consumer satisfaction influence the firm performance and stock returns; however, the related papers usually neglect the impacts of market status. This paper extends prior papers by considering the impacts of market status when exploring the relationship among innovation, consumer satisfaction, and firm performance. Empirical results show that in the bull markets innovation and consumer satisfaction do not significantly affect stock returns while in the bear markets stock returns are positively associated with the level of innovation and consumer satisfaction. These results suggest that managers should take market status into consideration when making marketing decisions.

We investigate the relationship between advertising and research and development (R&D) expenditures with the firm¡¦s systematic and unsystematic risks. Our data covers from January 1981 to December 2007 with more than two thousand publicly listed firms in the New York Stock Exchange. In addition to classical least squares approach, we utilize quantile regression model to examine whether the estimated slope parameters vary across different quantiles of the conditional distribution of the firm¡¦s systematic risk and unsystematic risk. We generate six empirical generalizations. (1) Advertising is significantly associated with lower systematic risk for firms with lower, median and higher systematic risk, but with no significant effects on the firms with extremely low systematic risk. (2) R&D is significantly associated with higher systematic risk for firms with median and higher systematic risk, with no significant effect for those with lower systematic risk. (3) Advertising is significantly associated with lower unsystematic risk for firms with higher unsystematic risk, but with no significant effects for those with median and lower unsystematic risk. (4) R&D is significantly associated with higher unsystematic risk for firms with median and higher unsystematic risk, with no significant effect for those with lower unsystematic risk. (5) Our evidence shows that both advertising and R&D have a stronger effect on firms with higher systematic risk (unsystematic risk) than on those with lower systematic risk (unsystematic risk). (6) Moreover, our evidence suggests that advertising and R&D tests resoundingly support our hypothesis that the coefficients vary across the quantiles.

Recently, there has been considerable concern with determining underpricing of initial public offerings (IPOs). This study utilizes both OLS and quantile regression model to examine whether pre-listing marketing expenditure reduce IPO underpricing using China A-share IPOs data. Our OLS result shows that firm¡¥s marketing expenditure could reduce IPO underpricing significantly that was consist with Luo¡¥s (2008) finding who investigate US IPOs market. With regard to quantile regression results, we find that pre-listing IPO marketing expenditures are significantly associated with lower underpricing for lower-underpricing stocks but with no significant effects for median-, and higher-underpricing stocks. We infer that: for lower-underpricing stocks, the risk premium investors require would be lowered because pre-listing marketing expenditures can help for raising transparency of the firm.

Several reports research the best prediction power of the credit risk models for different industries. The structural models use firm¡¦s information for firms¡¦ structural variables, such as asset value and asset volatility, to determine the time of default, but it suffer from some drawbacks, which represent the main reasons behind their relatively poor empirical performance. It require estimates for the parameters of the firm¡¦s asset value, which is nonobservable. Moody's KMV model is well known and useful among them, but it ignores recovery rate and difference in financial structure and industry. The reduced-form models fundamentally differ from typical structural models in the degree of predictability of the default. Reduced-form models use market data and assume the probability of default is exogenously generated. However, the basel committee for banking supervision proposed that risk is endogenous. The purpose of this paper is using quantile and threshold regression to introduce a new approach which is based on the Moody¡¦s KMV model, the Lu and Kuo ( 2005) and the Altman, Brooks Brady, Resti and Sironi (2005) to the evaluation of the endogenous probability of default and the endogenous recovery rate.

根據金管會銀行局的統計資料顯示，台灣在2005年2月底信用卡流通卡數已高逹44,611仟張，是1992年底信用卡流通卡數的近30倍。雖然信用卡流通卡數持續增長，在1992年底時成長率高逹62.1%，之後在這十年間信用卡流通卡數成長率幾乎都有30%以上的成長率，1996年成長率為48.7%，此時正為產品生命周期中的成長期。觀察近二年信用卡流通卡數的成長率，2004年只有16.7%，今年(2005年)成長率卻下滑到1%左右，可見信用卡市場已從生命周期中的成長期逐漸邁向成熟期。銀行若想在競爭激烈的信用卡市場中搶得先機，進而獲取利潤，應進行所謂產品的製程創新，即如何在信用卡進入產品生命周期的成熟期中，加強信用風險控管以降低成本、提高消費性產品即信用卡的品質和附加價值，以及如何進一步鞏固現有的信用卡客戶。本研究擬將提供一個具體之模型，以供日後銀行預測信用卡持卡人違約或剪卡之用。 本論文擬使用國內某家銀行在2004年3月底於資料倉儲中的客戶資料，有效分析客戶數共計128萬多筆。首先，本文先將信用卡客戶依人口統計變數、信用卡持卡人與發卡機構往來狀況、信用卡持卡人之使用狀況、信用卡持卡人之消費行為以及信用卡客戶付款狀況，探討信用卡客戶的剪卡概況。接著建構一個logistic model來預測客戶的剪卡機率，再用quantile regression model 分別對高剪卡率及低剪卡率之信用卡客戶進行分析。本文的重要發現有： 1. 年齡、是否使用循環利息在不同分量下，對於剪卡率的影響皆為負向關係，而且隨著分量愈大，剪卡率下降的幅度也愈多。 2. 每月限額、半年內交易次數、預借現金次數在不同分量下，對於剪卡率的影響皆為負向關係，而且隨著分量愈大，剪卡率下降的幅也愈少。 3. 婚姻狀況、有效信用卡數在不同分量下，對於剪卡率的影響皆為正向關係，而且隨著分量愈大，剪卡率增加的幅度也愈大。 銀行可根據重要的發現結果來制定授信政策，例如在每月限額部份，對於高剪卡率的客戶而言，若提高此客戶的信用額度，將使其剪卡率下降幅度少於低剪卡率的客戶，因此，銀行可著重在鞏固低剪卡率的客戶，藉由調高其信用額度，增加這群客戶對銀行信用卡的品牌忠誠度。或者可加以參考客戶的其它持卡消費行為，使授信政策更為完全，而且又可以滿足現存客戶的需求。