Scholarlink

This study explores the volatility dynamics and forecasting capabilities of the BSE SENSEX and the USD/INR exchange rate using the Generalized Autoregressive Conditional Heteroskedasticity (GARCH) (1,1) model. These two financial indicators serve as vital benchmarks for assessing economic stability and investor confidence in India. Utilizing daily data from April 2020 to March 2025a period marked by both post-pandemic recovery and global economic uncertaintythe research begins with descriptive statistics and stationarity testing via the Augmented Dickey-Fuller (ADF) method. ARIMA models are employed to estimate mean behaviour, while GARCH (1,1) captures time-varying volatility and persistence. The results reveal significant volatility clustering and long memory in both markets, with exchange rate fluctuations displaying slightly greater persistence. Diagnostic checks, including the Ljung-Box test, ARCH LM test, and Sign Bias test, confirm model adequacy. The findings suggest that GARCH-based models are effective tools for volatility forecasting and can aid investors, policymakers, and financial analysts in strategic risk assessment and decision-making.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Volatility Analysis and Forecasting of SENSEX and USD/INR Exchange Rate: An Empirical Study Using GARCH (1,1) Models

  • 章节

  • 发表日期：

    2026年

查看全文
  • Palak Chitlangiya,
  • Parul Bhatia,
  • Sudhi Sharma,
  • Mayank Goyal,
  • Shab Hundal

摘要

This study explores the volatility dynamics and forecasting capabilities of the BSE SENSEX and the USD/INR exchange rate using the Generalized Autoregressive Conditional Heteroskedasticity (GARCH) (1,1) model. These two financial indicators serve as vital benchmarks for assessing economic stability and investor confidence in India. Utilizing daily data from April 2020 to March 2025a period marked by both post-pandemic recovery and global economic uncertaintythe research begins with descriptive statistics and stationarity testing via the Augmented Dickey-Fuller (ADF) method. ARIMA models are employed to estimate mean behaviour, while GARCH (1,1) captures time-varying volatility and persistence. The results reveal significant volatility clustering and long memory in both markets, with exchange rate fluctuations displaying slightly greater persistence. Diagnostic checks, including the Ljung-Box test, ARCH LM test, and Sign Bias test, confirm model adequacy. The findings suggest that GARCH-based models are effective tools for volatility forecasting and can aid investors, policymakers, and financial analysts in strategic risk assessment and decision-making.