Figure - available from: International Journal of Financial Studies (IJFS)
This content is subject to copyright.
Source publication
Financial bubble prediction has been a significant area of interest in empirical finance, garnering substantial attention in the literature. This study aims to detect and forecast financial bubbles in the Vietnamese stock market from 2001 to 2021. The PSY procedure, which involves a right-tailed unit root test to identify the existence of financial...
Similar publications
Financial bubble prediction has been a significant area of interest in empirical finance, garnering substantial attention in the literature. This study aimed to detect and forecast financial bubbles in the Vietnamese stock market from 2001 to 2021. The PSY procedure, which involves a right-tailed unit root test to identify the existence of financia...
This study addresses the critical need for rapid and online measurement of liquid concentrations in industrial applications. Although the thermal lens effect (TLE) is extensively explored in laser systems for determining thermal lens focal lengths, its application in quantifying solution concentrations remains underexplored. This research explores...
Citations
... Ultimately, the use of AI-based predictive models allows for the identification of correlations between sentiment shifts and nonlinear price behavior. (Tran et al., 2023) argue that employing deep learning models to capture investor behavior during market crises yields more representative results compared to conventional statistical methods. ...
In recent years, retail investor participation in Southeast Asian capital markets has surged, contributing to increased market volatility and making sentiment analysis a critical factor in understanding price dynamics. This study investigates the relationship between social media sentiment and stock market fluctuations by focusing on Twitter data during periods of market volatility in Indonesia, Thailand, and Malaysia. The objective is to examine how collective investor emotions, as expressed through social media, correlate with daily stock index movements. Employing an exploratory quantitative approach, the study integrates Natural Language Processing (NLP) methods, both lexicon-based tools such as VADER and advanced transformer-based models like BERT and GPT, to classify over 150,000 tweets into positive, negative, and neutral sentiments. Sentiment scores were then aggregated and statistically tested using Pearson correlation with daily stock index returns, specifically the IDX Composite, SET Index, and FTSE Bursa Malaysia. The findings reveal a significant negative correlation between negative sentiment and market returns, particularly in the IDX Composite (r = -0.61, p < 0.05), indicating that pessimistic sentiment is associated with market downturns. Thailand’s SET Index and Malaysia’s FTSE Index showed moderate to weak negative correlations, with r = -0.43 and r = -0.27, respectively. These results highlight the sensitivity of emerging markets to emotionally driven retail behavior. The study concludes that AI-based sentiment analysis offers a valuable early warning tool for market volatility and can complement traditional financial indicators. It recommends developing AI-based sentiment dashboards and enhancing digital financial literacy to mitigate emotional reactivity among retail investors.
... Οι μέθοδοι μηχανικής μάθησης μπορούν να χειριστούν μεγάλα, πολύπλοκα σύνολα δεδομένων και να αποκαλύψουν κρυφά μοτίβα που μπορεί να χάνουν από τις παραδοσιακές στατιστικές μεθόδους. • Μάθηση χωρίς επίβλεψη: Οι αλγόριθμοι της συγκεκριμένης κατηγορίας δεν απαιτούν δεδομένα με ετικέτα και είναι ιδιαίτερα χρήσιμοι για τον εντοπισμό ανωμαλιών σε οικονομικά δεδομένα (Tran et al. 2023). Τεχνικές όπως η ομαδοποίηση (π.χ. ...
... Εντοπίζοντας ασυνήθιστα πρότυπα στις δραστηριότητες των δεικτών, τις ροές συναλλαγών των εταιριών ή στην κερδοφορία, οι επενδυτές μπορούν να διαφοροποιήσουν τις επενδύσεις απομονώνοντας τα χρηματοοικονομικά οχήματα που εμφανίζουν έντονες διακυμάνσειςανωμαλίες (Koyuncugil and Ozgulbas 2012;Vanini et al. 2023). Για παράδειγμα, κατά τη διάρκεια της χρηματοπιστωτικής κρίσης του 2007-9, τα ιδρύματα που χρησιμοποιούσαν ισχυρά συστήματα διαχείρισης κινδύνου βασιζόμενα σε ανίχνευση ανωμαλιών ήταν σε καλύτερη θέση για να μετριάσουν την επερχόμενη κατάρρευση (Tran et al. 2023). ...
Η παρούσα εργασία εκπονήθηκε στα πλαίσια του μαθήματος «Τεχνικές Ανάλυσης Δεδομένων» (DIS508_S24|01) του Διαπανεπιστημιακού Μεταπτυχιακού Προγράμματος «Ανάλυση Μεγάλων Δεδομένων και Χρηματοοικονομική Τεχνολογία», που διοργανώνεται από το Πανεπιστήμιο Νεάπολις Πάφου και το Ελληνικό Μεσογειακό Πανεπιστήμιο. Στην εργασία αυτή εξετάζεται η ανίχνευση ανωμαλιών στον χρηματοοικονομικό τομέα, με στόχο την ανάπτυξη μεθόδων και τεχνικών που θα επιτρέψουν την έγκαιρη αναγνώριση απρόβλεπτων και αποκλινουσών συμπεριφορών στις χρηματοπιστωτικές αγορές. Με την ανίχνευση ανωμαλιών επιτυγχάνεται η πρόληψη οικονομικών κρίσεων, η βελτίωση της διαχείρισης κινδύνου και η ενίσχυση της κανονιστικής συμμόρφωσης. Η εργασία αναλύει διάφορες μεθόδους ανίχνευσης ανωμαλιών, όπως στατιστικές μεθόδους, μεθόδους μηχανικής μάθησης και υβριδικές μεθόδους, και εξετάζει τις εφαρμογές τους στον χρηματοοικονομικό τομέα, περιλαμβάνοντας την ανίχνευση απάτης και τη διαχείριση κινδύνου.
Τα αποτελέσματα της μελέτης καταδεικνύουν τη σημασία της ανίχνευσης ανωμαλιών ως εργαλείο για τη διαμόρφωση ενός διαφανούς και σταθερού οικονομικού συστήματος. Παρά τις προκλήσεις που παρουσιάζονται, όπως η ποιότητα και διαθεσιμότητα των δεδομένων και η ερμηνευσιμότητα των μοντέλων, η επιτυχής ανίχνευση ανωμαλιών μπορεί να προσφέρει βελτιωμένα συστήματα έγκαιρης προειδοποίησης, καλύτερες πρακτικές διαχείρισης κινδύνου και αυξημένη κανονιστική συμμόρφωση. Η συνεχής έρευνα και η συνεργασία μεταξύ του ακαδημαϊκού κόσμου, της βιομηχανίας και των ρυθμιστικών αρχών είναι ζωτικής σημασίας για την πλήρη αξιοποίηση των δυνατοτήτων της ανίχνευσης ανωμαλιών στον χρηματοοικονομικό τομέα.
... Of course, this behavior further increases the price and patterns of the herding behavior of noise traders. Tran et al. (2023) define bubbles that are primarily driven by irrational investor behavior as classical. Ideally, rational arbitrageurs aim to exit the market just before the crash occurs. ...
... The primary focus of numerous studies concerns the identification of bubble conditions and volatility spillovers and undertaking crash forecasts (Lleo and Ziemba 2019;Yousaf and Hassan 2019). Tran et al. (2023) highlight that there is still a lack of a clear-cut definition of a bubble crash in the literature. Sornette (2009) defines crashes based on anomalous price patterns. ...
Hedging downside risk before substantial price corrections is vital for risk management and long-only active equity manager performance. This study proposes a novel methodology for crafting timing signals to hedge sectors’ downside risk. These signals can be integrated into existing strategies simply by purchasing sector index put options. Our methodology generates successful signals for price corrections in 2000 (dot-com bubble) and 2008 (global financial crisis). A key innovation involves utilizing sector correlations. Major price swings within six months are signaled when a sector exhibits high valuation alongside abnormal correlations with others. Utilizing the price-to-earnings ratio for identifying sectors’ high valuations is more beneficial than the bond–stock earnings yield differential. Our signals are also more efficient than those of standard technical analyses.
... The slow response of housing supply to market demand allows bubbles to grow and potentially burst, causing significant economic disruptions. Recent trends in housing markets in various global cities, marked by substantial price surges amid economic uncertainties, have renewed concerns about forming new bubbles and the effectiveness of existing strategies to address them [5][6][7]. Considering the significant risks associated with housing market bubbles and the limitations of current knowledge and policy tools, accurate prediction and effective management of these bubbles are crucial. 8 ...
... In our detailed evaluation, we compared our innovative algorithm against ten cutting-edge models: BankSAF [34], BankSAF [34], FiCAPS [36], RCPS [37], CuSeNet [38], GFRM [39], and MSP-LSTM [3], ML-VI [6], ML-BI [7], and LSTM-Autoencoder [10]. We also examined how our model performed without specific contributions, namely feature selection (FS), TLSTM, off-policy PPO, and hyperparameter optimization (HO), to gauge each contribution's impact on our model's overall effectiveness. ...
This research examines the link between the housing and stock markets, particularly emphasizing bubbles in the housing market. Typically, the fluctuations in the stock market have a more substantial effect on the movement of housing prices than the influence of housing market changes on the stock market. Nonetheless, when information from the housing market is used as an indicator, the changes in housing prices can serve as predictors of stock market volatility. Accordingly, we propose a novel early warning system (EWS) tailored to the housing market. This system is powered by an advanced spatial attention-based transductive Long Short-Term Memory (TLSTM) model, enhanced with an off-policy proximal policy optimization (PPO) algorithm for effective feature selection. The TLSTM model stands out for its ability to detect subtle temporal shifts in data, a significant improvement over conventional LSTM model. It emphasizes transductive learning that prioritizes data points near the test sample for increased accuracy. The off-policy PPO algorithm further strengthens our model by identifying complex patterns and relationships that might elude standard models, thereby improving the EWS’s predictive capabilities. Additionally, our methodology includes a finely tuned Differential Evolution (DE) algorithm that optimizes the model’s hyperparameters for peak performance. We validated our EWS using data from the Korean market, demonstrating its exceptional ability to forecast with an impressive accuracy of 85.422. By providing a more robust analytical framework, our study contributes significantly to the field of financial analytics, particularly in understanding and predicting the interactions between housing and stock markets during periods of market bubbles.
Finansal balonların öngörülmesi, uygulamalı finans literatüründe önemli bir araştırma konusu olarak kabul görmektedir. Bu çalışmanın amacı, 2002M01-2024M06 dönemi boyunca BİST 100, BİST Banka ve BİST Sınai endeksindeki finansal balonları ekonometrik ve makine öğrenimi yöntemlerini birleştirerek tespit etmek ve öngörmektir. Bu doğrultuda, ilk olarak balon dönemlerinin tespitinde Genelleştirilmiş Supremum Artırılmış Dickey-Fuller (GSADF) testi kullanılmıştır. Bu test, zaman serilerindeki yapısal kırılmaları ve volatilite değişimlerini dikkate alarak balonların tespitinde daha güçlü bir yöntem sunmaktadır. İkinci olarak, balon oluşumunu makroekonomik göstergelerle ilişkilendirerek tahmin etmek için Rastgele Orman Ağaçları (Random Forest Algorithm) yöntemi kullanılmıştır. Rastgele Orman (Random Forest) algoritması ile balon oluşum olasılığını etkileyen finansal ve makroekonomik faktörlerin göreceli önemi belirlenmiştir. Analiz sonuçları, Türkiye’de BİST 100, BİST Banka ve BİST Sınai endeksinde fiyat balonlarının oluşumunda para arzı, döviz kuru, tüketici fiyat endeksi, FED faiz oranı, Bileşik Öncü Göstergeler Endeksinin belirleyici olduğunu göstermektedir. Bu doğrultuda, balon risklerini azaltmak ve finansal istikrarı sağlamak için para politikası ve makro ihtiyati tedbirlerin etkin kullanımının önemini ortaya koymaktadır. Aynı zamanda makine öğrenmesi yöntemlerinin finansal balonların erken tespiti ve piyasa risklerinin yönetimi açısından önemli bir potansiyel taşıdığını göstermektedir. Modelin performansını değerlendiren sonuçlar, önerilen yaklaşımın yüksek tahmin gücüne sahip olduğunu ve finansal balonların öngörülmesinde etkili bir alternatif olabileceğini göstermektedir.
It is known that there is a phenomenon in the economy when the basic value of any tangible or intangible asset differs significantly from its market value, and the growing demand causes an increase in prices. Thus, trading in significant volumes of such assets is carried out at an inflated price, which in turn creates financial bubbles. The theory of efficient markets, present in information sources, states that the available asset prices are always in line with market requirements and take everything into account, but at the same time, it is assumed that market participants act rationally when making their decisions. However, markets are ruled by people, and as sociological studies show, all people are irrational to one degree or another in their actions. To date, there are no clear and effective tools that allow predicting and preventing the formation of financial bubbles with sufficient accuracy, and experts offer expert judgments about the risks of a bubble based on the analysis of financial time series and the comparison of the expected market with the available data of previous crisis situations. The work offers an overview of information sources, which consistently present the essence and prerequisites of the appearance of financial bubbles, the process of their formation, their impact on economic indicators both at the global level and on the economy of individual countries. Also, the considered works describe in detail the behavioral model of the financial bubble, the supply and demand function, the general equation of the asset price, and the dynamics of behavioral contagion of the market population. An overview of the behavioral mathematical model of the financial bubble is presented separately. The article argues the relevance of creating informational methods for forecasting the emergence of financial bubbles, as well as building a model capable of demonstrating trends, not only of ordinary bubbles, but also of serial bubbles. Using the Python programming language, auxiliary libraries and frameworks, with the application of the indicated behavioral model of the financial bubble, data simulation was carried out and a number of interactive visualizations of the process of their formation, development and disappearance were built. Also, the obtained results make it possible to assert that people, even without professional knowledge in the field of information technologies and programming, can create sufficiently productive information systems for analyzing financial market data.
Стаття є оглядово-інформаційним викладенням матеріалу з використання моделей машинного навчання для прогнозування фінансових показників з можливістю покращення ухвалення рішень з метою формування збалансованого портфеля акцій. Об’єднання глибокого аналізу з машинним навчанням, використовуючи методи лінійної регресії, дерев рішень, випадкових лісів та платформ автоматичного навчання, є потужним інструментом для передбачення ринкових трендів та стратегічних рухів, а також тестування гіпотез. Особлива увага приділяється аналізу ефективності різних методів машинного навчання в умовах невизначеності та турбулентності ринку. Використання моделей штучного інтелекту, дозволяє враховувати залежності у фінансових даних. На основі проведених досліджень визначаються основні напрями подальших дій у цій сфері, включаючи впровадження інформаційних моделей у практичну діяльність фінансових інститутів.
This study meticulously examines market bubbles within specific sectors of the National Stock Exchange (NSE) over the period from January 2017 to December 2023, employing robust methodologies like RADF, SADF, and GSADF tests. The analysis, centered on 11 sectoral indices, integrates GSADF values with RADF and SADF, offering nuanced perspectives that underscore the sector-specific nature of bubbles. Notably, the study highlights bubble occurrences during the 2020 global crisis due to pandemic, emphasizing their dynamic and diverse manifestations amid the pandemic. Exclusive identification of bubbles in NSE IT, NSE Metal, and NSE Pharma enriches the strategic insights available to investors, facilitating informed decision-making and risk management. The sector-wise approach contributes to a holistic understanding of market dynamics, providing investors with valuable tools to navigate the intricacies of the financial landscape. Future research avenues may delve into regulatory impacts on sector-specific bubbles and explore the interplay between macroeconomic indicators and sectoral bubbles, offering deeper insights into market dynamics.
