Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

The Forecasting of Daily Stock Returns with Sentiment Analysis and Google Trends

Zadrazil, Pavel LU (2019) EKHS51 20191
Department of Economic History
Abstract
Previous research has shown a link between investor sentiment and stock market movement, as well as the correlation of search engine queries and market volatility. This research aims to extend these findings by predicting the exact firm-level daily return of Apple and Amazon. The daily returns are predicted with an artificial neural network model. In theory, this model is able to learn similarly to a human brain, it finds the relevance of individual inputs (investor sentiment and investor interest to trade) and then computes daily returns based on the inputs’ significance. Even though the model used in this research has an ability to learn, it fails to predict exact daily returns. However, the model is able to predict the positive or... (More)
Previous research has shown a link between investor sentiment and stock market movement, as well as the correlation of search engine queries and market volatility. This research aims to extend these findings by predicting the exact firm-level daily return of Apple and Amazon. The daily returns are predicted with an artificial neural network model. In theory, this model is able to learn similarly to a human brain, it finds the relevance of individual inputs (investor sentiment and investor interest to trade) and then computes daily returns based on the inputs’ significance. Even though the model used in this research has an ability to learn, it fails to predict exact daily returns. However, the model is able to predict the positive or negative direction of the stock returns fairly accurately. (Less)
Please use this url to cite or link to this publication:
author
Zadrazil, Pavel LU
supervisor
organization
alternative title
Exploring the Apple and Amazon Case
course
EKHS51 20191
year
type
H1 - Master's Degree (One Year)
subject
keywords
Predict, Equity Market, Daily Returns, Artificial Neural Network, Investor Sentiment, Investor Interest
language
English
id
8991298
date added to LUP
2019-08-22 08:22:39
date last changed
2019-08-22 08:22:39
@misc{8991298,
  abstract     = {{Previous research has shown a link between investor sentiment and stock market movement, as well as the correlation of search engine queries and market volatility. This research aims to extend these findings by predicting the exact firm-level daily return of Apple and Amazon. The daily returns are predicted with an artificial neural network model. In theory, this model is able to learn similarly to a human brain, it finds the relevance of individual inputs (investor sentiment and investor interest to trade) and then computes daily returns based on the inputs’ significance. Even though the model used in this research has an ability to learn, it fails to predict exact daily returns. However, the model is able to predict the positive or negative direction of the stock returns fairly accurately.}},
  author       = {{Zadrazil, Pavel}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{The Forecasting of Daily Stock Returns with Sentiment Analysis and Google Trends}},
  year         = {{2019}},
}