Market crash factors and developing an early warning system: Evidence from Asia

  • 397 Views
  • 119 Downloads

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License

Market crashes pose significant risks to the stability and performance of financial markets, making the development of an early warning system crucial. This study utilizes exchange rate volatility and investor sentiment to predict market crashes. While several studies have examined factors affecting market crashes in developing countries. This study aims to develop an early warning system for investors to minimize investment risk using Exchange Rate Volatility and Investor Sentiment. The study focused on seven countries: Indonesia, Malaysia, Singapore, the Philippines, Thailand, Vietnam, and Mongolia. The stock exchanges examined included Jakarta Stock Exchange Composite, FTSE Malaysia KLCI, FTSE Singapore, SET Index, PSEi, HNX/HNXI, and MNE Top 20/MNETOP20. The analysis involved assessing early warning systems to provide valuable supplementary information for decision making and evaluating market vulnerabilities. The logistic regression equation was utilized to model market crashes, incorporating variables such as exchange rate volatility and investor sentiment while considering their interactions as moderating factors. The results indicate that exchange rate volatility and investor sentiment have a significant negative effect on market crashes, with probabilities of 0.0082 and 0.000 Furthermore, investor sentiment acts as a mediator for exchange rate volatility, amplifying its impact on market crashes. This suggests that higher exchange rate volatility and negative investor sentiment increase the likelihood of market crashes. Exchange rate volatility and investor sentiment can serve as early warning indicators, emphasizing the importance of monitoring these factors for market participants and policymakers.

view full abstract hide full abstract
    • Figure 1. Research framework
    • Table 1. Operational variables
    • Table 2. Descriptive analysis
    • Table 3. Correlation matrix
    • Table 4. Matrix of constant Markov transition probabilities
    • Table 5. Binary logit regression output
    • Data curation
      Lisa Kustina, Rachmat Sudarsono, Nury Effendi
    • Formal Analysis
      Lisa Kustina, Rachmat Sudarsono, Nury Effendi
    • Funding acquisition
      Lisa Kustina
    • Investigation
      Lisa Kustina, Rachmat Sudarsono, Nury Effendi
    • Methodology
      Lisa Kustina, Rachmat Sudarsono, Nury Effendi
    • Project administration
      Lisa Kustina
    • Resources
      Lisa Kustina, Rachmat Sudarsono, Nury Effendi
    • Software
      Lisa Kustina
    • Visualization
      Lisa Kustina
    • Writing – original draft
      Lisa Kustina
    • Writing – review & editing
      Lisa Kustina, Rachmat Sudarsono, Nury Effendi
    • Conceptualization
      Rachmat Sudarsono, Nury Effendi
    • Supervision
      Rachmat Sudarsono, Nury Effendi
    • Validation
      Rachmat Sudarsono, Nury Effendi