This study explores the role of the asymmetrical influence of stock market reaction to the fluctuations in US dollars against the domestic currencies of the South Asian region. The symmetrical Panel based autoregressive distributed lag model (PARDL) model and panel-based non-linear autoregressive distributed lag model (NARDL) model with the Pooled Mean Group Approach (PMG) approach were employed to estimate the impact of stock market bullish and bearish behaviour on local currency variabilities of the South Asian region. The results show that during the pre-economic recession regime, negative fluctuations in stock indexes had a direct association with exchange rate fluctuations for a longer period. However, after the crisis, both negative and positive stock index shocks have formulated an indirect or inverse association with exchange rate fluctuations. These findings pose important practical implications for the short-term and long-term shareholders and financial analysts. Hence, it is recommended that shareholders and financial analysts should base their investment in South Asian stock indices on negative and positive shocks since the non-linear connectivity between both variables also exhibits regime dependence characteristics.