Trend-Following Indicators Under Stable and High-Volatility Market Regimes: An Empirical Assessment Using GARCH-Based Volatility Classification

Abstract

This study investigates whether the directional accuracy of commonly used technical indicators changes with market volatility and whether combining several indicators can improve signal reliability. The analysis draws on 500 daily financial observations and considers moving averages, the Relative Strength Index (RSI), the Moving Average Convergence Divergence (MACD), the stochastic oscillator and Bollinger Bands. Conditional volatility is estimated with a GARCH(1,1) model. Observations are then divided into stable and high-volatility regimes depending on whether estimated conditional variance falls below or above its sample mean, resulting in 260 stable and 240 high-volatility observations. Indicator accuracy is measured by comparing conventional buy and sell signals with the direction of the following day’s price movement. The two regimes display noticeably different volatility patterns. Return volatility rises from 1.213% in stable conditions to 2.489% in the high-volatility regime, while estimated volatility persistence (α + β) increases from 0.890 to 0.983. At the aggregate level, individual indicators generally achieve directional accuracy close to 50%. Combining indicators produces higher observed accuracy, although this improvement comes with a substantial reduction in the number of signals. The combination of RSI, Bollinger Bands and MACD records the highest observed accuracy at 66.7%, corresponding to 10 correct predictions out of only 15 signals. Given this limited number of observations, the result should be viewed cautiously and does not establish robust predictive superiority. The study adds to the existing literature by examining technical-signal reliability explicitly in relation to volatility conditions and by comparing individual indicators with confirmation-based strategies within the same empirical setting. The findings indicate that prevailing volatility conditions matter when interpreting technical signals and that fixed technical rules may not perform consistently across market environments. Since the analysis concerns directional accuracy rather than realised returns, the findings should not be interpreted as evidence of trading profitability. Further research using out-of-sample validation is needed to assess the economic relevance and stability of these results.