This study focuses on the future orientation and sentiment of investors in relation to Turkey and its stock market. First, I analyze the future orientation of investors, and their polarity, by constructing Twitter-based sentiment measures and relate them to existing Consumer Confidence Indexes and Business Confidence Indexes. Then, I estimate the relation of these measures with the Istanbul Stock Market Index (BIST) to check whether the sentiment and future orientation of investors has affected the stock market in Turkey from 2010 to 2019. Twitter-based sentiment measures are created using a sentiment dictionary-based approach, applied to a filtered collection of tweets about Turkey, published from 2010 to 2019. In particular, I use only tweets that include future-oriented semantic markers and tweets that have to do with investing or doing business in Turkey. Constructed Twitter-based sentiment measures can be used to understand the sentiment dynamics and expectations polarity w.r.t investing in the Turkish market in the upcoming periods. Being based on future-oriented tweets, the proposed measures signal the degree of optimism or pessimism conveyed through Twitter. By analyzing the fluctuations of these measures over time, I show the sensitiveness of the international audience to key events occurring in Turkey and their relationship with financial market fluctuations.

Twitter-Based Future Orientation Measures and The Turkish Stock Market

Atabay, Nil
2020/2021

Abstract

This study focuses on the future orientation and sentiment of investors in relation to Turkey and its stock market. First, I analyze the future orientation of investors, and their polarity, by constructing Twitter-based sentiment measures and relate them to existing Consumer Confidence Indexes and Business Confidence Indexes. Then, I estimate the relation of these measures with the Istanbul Stock Market Index (BIST) to check whether the sentiment and future orientation of investors has affected the stock market in Turkey from 2010 to 2019. Twitter-based sentiment measures are created using a sentiment dictionary-based approach, applied to a filtered collection of tweets about Turkey, published from 2010 to 2019. In particular, I use only tweets that include future-oriented semantic markers and tweets that have to do with investing or doing business in Turkey. Constructed Twitter-based sentiment measures can be used to understand the sentiment dynamics and expectations polarity w.r.t investing in the Turkish market in the upcoming periods. Being based on future-oriented tweets, the proposed measures signal the degree of optimism or pessimism conveyed through Twitter. By analyzing the fluctuations of these measures over time, I show the sensitiveness of the international audience to key events occurring in Turkey and their relationship with financial market fluctuations.
2020-07-28
File in questo prodotto:
File Dimensione Formato  
877919-1254969.pdf

non disponibili

Tipologia: Altro materiale allegato
Dimensione 2.57 MB
Formato Adobe PDF
2.57 MB Adobe PDF   Richiedi una copia

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14247/16980