Please use this identifier to cite or link to this item: https://scholarbank.nus.edu.sg/handle/10635/147768
Title: STATISTICAL ANALYSIS OF SEASONALITY IN CROSS-SECTIONAL STOCK RETURNS
Authors: LIU YANAN
Issue Date: 2014
Citation: LIU YANAN (2014). STATISTICAL ANALYSIS OF SEASONALITY IN CROSS-SECTIONAL STOCK RETURNS. ScholarBank@NUS Repository.
Abstract: In the past decade, significant amount of research has been devoted to trend analysis of stock returns and predictability of the market. Many results have shown that stocks exhibit relatively stable periodicity for up to the past 20 years, and the seasonality trend is found to be pervasive across the globe (Jegadeesh N., 1990; Jegadeesh & Titman, 1993; Griffin, Ji, & Martin, 2003; Das & Rao, 2011). However, the underlying causes are yet to be discovered. Existing literature shows that some commonly used finance measurements, such as types of industry, dividends, earnings announcements and fiscal year are not able to explain the periodicity (Heston & Sadka, 2008). Building on previous research, this project hopes to examine and understand the causes of the seasonality pattern in the cross-sectional stock returns. In this dissertation, robustness of the seasonality in cross-sectional stock returns is tested. Regression analysis is run on stock data from NYSE, AMEX and NASDAQ across different time spans. A detailed case analysis of annually-lagged returns is conducted. It is found that the seasonality persists when database is modified, which confirms the robustness of the trend. To understand the underlying drivers, technical derivation of regression equations is performed and the cross-sectional regression estimate is decomposed into three components, namely the time-series regression estimate, bias and error, for further analysis. It is found that both bias and error terms are negative, implying under-estimated magnitude of the seasonality in the previous cross-sectional regression analysis.
URI: http://scholarbank.nus.edu.sg/handle/10635/147768
Appears in Collections:Bachelor's Theses

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
b33792677.pdf1.11 MBAdobe PDF

RESTRICTED

NoneLog In

Page view(s)

12
checked on Jul 3, 2020

Download(s)

6
checked on Jul 3, 2020

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.