Kiekvienas investuotojas susiduria su efektyvių investicinių sprendimų priėmimo problema. Yra daug metodų, kuriais stengiamasi išanalizuoti finansų rinkoje vykstančių pokyčių priežastis bei remiantis tokia informacija numatyti ateities tendencijas. Vienas iš būdų yra investuotojų sentimentų prognozavimas. Šio straipsnio tyrimo tikslas yra atlikti skirtingų investuotojų sentimentų prognozavimą ir įvertinti prognozavimui naudojamo modelio patikimumą, t. y. siekiama atrasti patikimą sentimentų prognozavimo algoritmą. Tyrimui naudojamas dirbtinio intelekto giliojo mokymosi ilgos trumpalaikės atminties (LSTM) tinklų algoritmas bei grafinis gautų rezultatų vaizdavimas. Atlikus tyrimą buvo pastebėta, kad kiekvienu sentimentų prognozavimo atveju gauta paklaida (RMSE) buvo labai maža, o tai reiškia, kad prognozavimui naudojamas algoritmas yra labai patikimas. Sentimentų prognozavimas kartu su racionaliais prognozavimo metodais gali papildyti prekybos strategiją ar paramos sistemą investuotojui.
Purpose – The paper analyses two different paradigms of investor behaviour that exist in the financial mar-ket – the herding and contrarian behaviour. The main objective of the paper is to determine which pattern of investor behaviour better reflects the real changes in the prices of financial instruments in the financial markets. Research methodology – Algorithms of technical analysis, deep learning and classification of sentiments were used for the research; data of positions held by investors were analysed. Data mining was performed using “Tweet Sentiment Visualization” tool. Findings – The performed analysis of investor behaviour has revealed that it is more useful to ground financial decisions on the opinion of the investors contradicting the majority. The analysis of the data on the positions held by investors helped to make sure that the herding behaviour could have a negative impact on investment results, as the opinion of the majority of investors is less in line with changes in the prices of financial instruments in the market. Research limitations – The study was conducted using a limited number of investment instruments. In the future, more investment instruments can be analysed and additional forecasting methods, as well as more records in social networks can be used. Practical implications – Identifying which paradigm of investor behaviour is more beneficial to rely on can offer ap-propriate practical guidance for investors in order to invest more effectively in financial markets. Investors could use investor sentiment data to make practical investment decisions. All the methods used complement each other and can be combined into one investment decision strategy. Originality/Value – The study compared the ratio of open positions not only with real price changes but also with data obtained from the known technical analysis, deep learning and sentiment classification algorithms, which has not been done in previous studies. The applied methods allowed to achieve reliable and original results.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.