Log in
|
Help
|
English

English

Engish-Pirate

한국어

Search

Advanced Search

Find results with:

Add another field

Search by date

to

Searching collections:

CCSU Theses and Dissertations

Add or remove collections

Home Using data mining to model market reaction to corporate earning announcements

Reference URL

Share

Add tags

Comment

Rate

To link to this object, paste this link in email, IM or document

To embed this object, paste this HTML in website

Using data mining to model market reaction to corporate earning announcements

View Description

Download

small (250x250 max)

medium (500x500 max)

large ( > 500x500)

Full Resolution

Print

you wish to report:

Your comment:

Your Name:

Submit

Cancel

...

Back to top

Powered by CONTENTdm®

Select the collections to add or remove from your search

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

R

S

T

U

V

W

X

Y

Z

Select All Collections

C

CCSU Student Publications

CCSU Theses and Dissertations

G

GLBTQ Archives

M

Modern Language Oral Histories

O

O'Neill Archives Oral Histories

P

Polish American Pamphlets

T

Treasures from the Special Collections

V

Veterans History Project

OK

Cancel

Identifier	Thesis 2241
Author	Gu, Judith Y., 1970-
Title	Using data mining to model market reaction to corporate earning announcements
Publisher	Central Connecticut State University
Date	2012
Resource Type	Master's Thesis
Notes	Investor community and stock market watch companies' earnings report closely as earnings report is the most essential fundamental barometer of how financially healthy a company is. All public companies announce earnings every quarter, and the stock market scrutinizes their revenue (top line), net income (bottom line) and future economic guidance. Investors also read into fine details including words and tones that management uses at the earnings conference call. These numbers are immediately compared to "market consensus". If a company misses consensus, it most likely will underperform its peers and general market massively. If the company beats consensus, it will most likely outperform. However, to profit from earnings events is not as simple as "to buy the winners and sell the losers". To begin with, prices make "gap move" on the "earnings move day" - i.e. if the company misses earnings significantly, it may drop 10% on the open and leave investors no opportunity to sell the company and profit from the first 10% move. On the other hand, if the company beats consensus, price likely will jump before market opens and make it very expensive to buy the winners. The question now becomes: given the stock has already made big moves on earnings day, should investors keep chasing the momentum or should they bet on a reversal of the big move hoping that the market has overreacted? This paper is going to elaborate on a strategy that buys and sells shares around their earnings - a long / short portfolio to trade "earnings surprises". The method employed by this research is to define what constitutes "earnings surprise" - the trigger of putting on a trade. It then conducts a comprehensive back test trading strategy for 3,000 companies' earnings events over the last 4 years to prove validity of the strategy. Optimization from different factors that might have impacted the strategy will be analyzed in detail and CART analysis from R statistical software package will be applied to find out the factor that has the biggest impact. The conclusion is that there is a short term momentum drift after earnings events and trading the drift systematically can yield a profitable return
Subject	Data mining Finance -- Mathematical models
Department	Department of Mathematical Sciences
Advisor	Larose, Daniel T.
Type	Text
Digital Format	application/pdf
Language	eng
Rating

Find results with:

Searching collections:

Home Using data mining to model market reaction to corporate earning announcements

Using data mining to model market reaction to corporate earning announcements

Description

Tags

Comments