Publications
Detailed Information
Regression with Partially Observed Ranks on a Covariate: Distribution-Guided Scores for Ranks : 부분적으로 관측된 순위 공변량을 이용한 회귀분석: 순위에 대한 분포-유도 스코어 함수
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 임요한 | - |
dc.contributor.author | 김윤응 | - |
dc.date.accessioned | 2017-07-14T00:32:04Z | - |
dc.date.available | 2017-07-14T00:32:04Z | - |
dc.date.issued | 2016-08 | - |
dc.identifier.other | 000000136666 | - |
dc.identifier.uri | https://hdl.handle.net/10371/121163 | - |
dc.description | 학위논문 (박사)-- 서울대학교 대학원 : 통계학과, 2016. 8. 임요한. | - |
dc.description.abstract | This work is motivated by a hand-collected data set from one of the largest
Internet portals in Korea. This data set records the top 30 most frequently discussed stocks on its on-line message board. The frequencies are considered to measure the attention paid by investors to individual stocks. The empirical goal of the data analysis is to investigate the effect of this attention on trading behavior. For this purpose, we regress the (next day) returns and the (partially) observed ranks of frequencies. In the regression, the ranks are transformed into scores, for which purpose the identity or linear scores are commonly used. In this thesis, we propose a new class of scores (a score function) that is based on the moments of order statistics of a random variable Z. The new scores are shown to be flexible in modeling the desired features (e.g., monotonicity or convexity) of the scores. In addition, if the true covariate X is drawn from a location-scale family and Z is its standardized distribution, then the least-squares estimator calculated using the proposed scores consistently estimates the true correlation between the response and the covariate and asymptotically approaches the normal distribution. We also propose a procedure for diagnosing a given score function and selecting one that is better suited to the data. We numerically demonstrate the advantage of using a correctly specifed score function over that of the identity scores (or other misspecifed scores) in estimating the correlation coefficient. Finally, we apply our proposal to test the effects of investors' attention on their returns using the motivating data set. | - |
dc.description.tableofcontents | Chapter 1.Introduction 1
Chapter 2.Literature Review 5 2.1.Score-based Analysis 6 2.1.1.Simple linear rank statistics 6 2.1.2.Two-way ANOVA model 7 2.2.Choice of Score Function 9 2.2.1.2×K contingency table 9 2.2.2.Drawbacks of integer scoring 11 Chapter 3.Distribution-Guided Scores for Ranks 12 3.1.Relationship between score and quantile function 12 3.2.The moment problem of the order statistics 13 3.3.Features of location-scale family assumption 15 Chapter 4.Simple Linear Regression 16 4.1.Least-Squares Estimator 17 4.2.Residual Analysis 23 4.3.An Estimator with Unranked Observations 24 Chapter 5.Numerical Study 27 5.1.Study setup 27 5.2.Interpretations and results 28 Chapter 6.Data Examples 32 6.1.Data Description 32 6.2.Attention and Predictive Stock Returns 34 6.3.Regression with Ranks 35 6.4.Test of the Effect of Investor Attention on the Next-day Returns 38 6.5.Test on overall Correlation 41 Chapter 7.Concluding remarks 46 Bibliography 49 국문 초록 53 | - |
dc.format | application/pdf | - |
dc.format.extent | 693924 bytes | - |
dc.format.medium | application/pdf | - |
dc.language.iso | en | - |
dc.publisher | 서울대학교 대학원 | - |
dc.subject | Concomitant variable | - |
dc.subject | investor attention | - |
dc.subject | linear regression model | - |
dc.subject | moments of order statistics | - |
dc.subject | partially observed ranks | - |
dc.subject | scores of ranks. | - |
dc.subject.ddc | 519 | - |
dc.title | Regression with Partially Observed Ranks on a Covariate: Distribution-Guided Scores for Ranks | - |
dc.title.alternative | 부분적으로 관측된 순위 공변량을 이용한 회귀분석: 순위에 대한 분포-유도 스코어 함수 | - |
dc.type | Thesis | - |
dc.description.degree | Doctor | - |
dc.citation.pages | 54 | - |
dc.contributor.affiliation | 자연과학대학 통계학과 | - |
dc.date.awarded | 2016-08 | - |
- Appears in Collections:
- Files in This Item:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.