The Impact of Global Stock Index Fluctuations on U.S. Inflation: An Empirical Analysis Based on ARIMA and VAR Models

Abstract

This study investigates the influence of major global stock indices on U.S. inflation, incorporating both domestic indicators (S&P 500 and NASDAQ Composite) and international benchmarks (Hong Kong Hang Seng Index and German DAX). Using monthly data spanning over two decades, we employ Autoregressive Integrated Moving Average Models (ARIMA models) to identify time series characteristics and Vector Autoregressive Model (VAR models) to analyze dynamic interrelationships. The U.S. Dollar Exchange Rate Index is included as a control variable to account for currency effects. Results indicate that both inflation and stock index series are non-stationary but achieve stationary after first-order differencing. Empirical analysis shows that lagged returns—especially from two months prior—of all major indices exhibit a significant positive relationship with U.S. inflation, while the exchange rate exerts a more complex, lag-dependent influence. Granger causality tests suggest that certain indices, particularly the NASDAQ and DAX, contain predictive information for future U.S. inflation movements, whereas feedback from inflation to equity markets is minimal. Impulse response analysis further reveals that unexpected shocks to stock indices lead to short-term increases in inflation, peaking within five to ten months before fading. These findings underscore the important role of global equity markets in shaping U.S. inflationary trends.