Frontiers in Business, Economics and Management ISSN: 2766-824X | Vol. 17, No. 1, 2024 96 Research on the Dynamic Relationship Between Asset Price Fluctuation and Foreign Exchange Market Kexin Zhang College of Business and Economics, Australian National University, Acton, ACT, 2601, Australia Abstract: Through theoretical discussion and empirical investigation, this study aims to elucidate the dynamic interconnection between asset price shocks and foreign exchange, and analyze the nature of their interaction effects and their substantive economic consequences. Based on the financial data of recent years, with the help of time series analysis method and vector autoregressive model (VAR) tools, this paper confirms the significant disturbance of asset value changes on foreign exchange rates and the lasting characteristics of this effect. It is found that asset volatility has a decisive short-term impact on the currency exchange market, and this correlation shows the characteristics of travel alienation under the condition of variable elements. Conversely, the evolution of the foreign exchange environment also affects asset pricing, building a two-way feedback chain. This study not only deepens the existing theoretical framework, but also provides strategic insights for regulators to cope w ith the shock of property values and ensure the smooth operation of foreign exchange. At the same time, this paper also carefully discriminates its own limitations and points out the possible ways of follow-up research. Keywords: Asset price volatility, Foreign exchange market, Dynamic relationship, Market linkage effect, Policy advice. 1. Introduction As global economic integration accelerates, financial markets worldwide become more interconnected. Asset price volatility, involving stocks, bonds, and commodities, gains global attention. These fluctuations reflect investors' economic expectations and may pose systemic risks. The foreign exchange market, crucial for international capital flows, directly impacts trade and investment. Exploring the link between asset price volatility and the foreign exchange market is vital for understanding the global economy. Knowing how they mutually affect each other is crucial for investors and policymakers. Studies show complex interactions, with some scholars discussing theoretical impacts and others empirically testing this relationship [1-2]. However, most studies focus on specific periods or countries, lacking a global perspective, and results are limited by data constraints. This research can use comprehensive theoretical analysis and empirical investigation methods to reveal the interaction effect between asset value shocks and foreign exchange markets in a globalized environment. The first task focused on elucidating the mechanisms by which asset price movements affect exchange rate markets. Second, it provides a deeper understanding of how FX market dynamics feed back into the asset price system. Based on these insights, we will formulate targeted policy guidance. 2. Theoretical Framework Financial markets interact, with asset price fluctuations and foreign exchange rates showing significant correlation [3]. Securities market movements reflect macroeconomic health, influencing the foreign exchange market. A stock market boom suggests economic expansion, boosting the domestic currency. Equity declines may lead to seeking external safe- havens, lowering local currency values. The foreign exchange sector also impacts asset prices. A strong currency benefits domestic consumers by reducing imports' cost but pressures export-oriented businesses. Conversely, a weak currency aids exports but burdens companies with higher import costs. 2.1. Exchange Rate Determination Theory The principle of purchasing-power parity states that a healthy exchange rate should reflect the value of goods and services exchanged for in each currency. According to this theory, in the long run, exchange rates should be calibrated to ensure that price equivalence remains constant across countries. In the short run, however, the real dynamics of the real exchange rate tend to diverge from the PPP, which can be attributed to the complex effects of non-tradable transactions and market expectations. The interest rate parity hypothesis states that the interest rate differential between two currencies should match the forex market's expectations of future exchange rate changes, further eliminating the scope for cross-border risk-free arbitrage. This theory highlights the correlation between interest rate differentials and expected exchange rate movements. The monetarist perspective focuses on the idea of how imbalances between the supply and demand of money shape exchange rates. According to this model, if the growth of a country's money supply exceeds its economic expansion rate, the value of its currency will tend to depreciate [4]. The core argument of the monetarist model is the dominant role of monetary policy in exchange rate fluctuations. A new behavioral finance perspective suggests that in addition to economic fundamentals, traders' psychological situations, emotional tendencies and decision- making biases can also have significant disturbing effects on exchange rates. 2.2. Asset Pricing Model The Capital Asset Pricing Model (CAPM) assesses expected returns on risky assets based on systemic risk (Ξ² coefficient) [5]. It assumes rational investors, efficient markets, and quick price corrections. Arbitrage Pricing Theory (APT) is a multi-factor model considering macroeconomic factors like inflation and interest rates, offering flexible risk management. The Fama-French three- 97 factor model adds firm size and book-to-market ratio to CAPM, arguing higher returns for small-cap and high book- to-market stocks due to higher risk. 2.3. Research Hypothesis From the foregoing theoretical discussion, we can conceive the following hypothesis. H1. The volatility of securities prices has a significant disturbing effect on the foreign exchange field. H2. The dynamic changes in the foreign exchange field exert an important counterforce to the fluctuations of asset values. 3. Methodology 3.1. Data Source and Sample Selection The study used data from internationally recognized financial databases such as Bloomberg, Wind and others. The data covers daily closing prices over selected time periods, including but not limited to stock indexes, bond yields, commodity prices, and exchange rates of major currency pairs. The sample includes major financial indicators from multiple countries and regions to ensure that the findings are broadly representative and generalizable. A long time span was selected during the sample period in order to capture market performance in different economic cycles. 3.2. Variable Definition The core changes in the foreign exchange market, reflected in exchange rate dynamics, are the dependent variable. Independent variables include financial market value swings, such as stock market, bond yield, and commodity pricing fluctuations. Control variables, like GDP growth, inflation, policy interest rate adjustments, and investment climate, are introduced to eliminate distractions. 3.3. Research Method By calculating the mean value, standard deviation, extreme value and other core statistical indicators of each variable, we can gain insight into the distribution characteristics of the data and further provide a solid preliminary preparation for the subsequent in-depth analysis. In order to investigate the existence of unit root phenomenon in each time series, we propose to adopt the enhanced Dickey-Fuller (ADF) test and the Kwiatkowski-Phillips-Schmidt-Shin (KPSS) test. This step is crucial for evaluating the stability of variables. Once the instability of time series is revealed in the unit root test, we will then carry out the Engle-Granger two-step method or Johansen co-integration test, so as to explore whether there is a durable linear equilibrium correlation between these unsteady variables. We use vector autoregressive (VAR) model to explore the dynamic interaction between asset price volatility and foreign exchange market. The VAR model has the unique advantage of taking into account the interdependence among multiple time series and estimating the short - and long-term interaction between variables without presupposing clear causal links. Using Granger causality test, we will identify whether asset price fluctuations constitute the "anthems" of changes in the forex market, and vice versa. 3.4. Empirical Design Based on the methods and strategies mentioned, we will build a vector autoregression (VAR) model using detailed sample data for accurate parameter estimates. To ensure the model's theoretical and practical effectiveness, we will rigorously verify and optimize it. This includes conducting a systematic residual analysis, focusing on white noise testing to ensure residuals are purely random errors. We will also use visual tools like autocorrelation graphs (ACF) and partial autocorrelation graphs (PACF) to analyze residuals, checking for periodic or trend components and identifying overfitting or missing variables. These tests will verify that residuals meet requirements of no autocorrelation, constant mean square error, and normal distribution. 4. Empirical Analysis 4.1. Descriptive Statistical Analysis Table 1 shows the mean value, standard deviation, minimum value, maximum value and other statistics of each major variable, which helps us to preliminarily understand the distribution of data. Table 1. Experimental results The observations revealed in this study suggest that there is a unique dynamic correlation between stock market index, securities trading profits, futures commodity valuation fluctuations and exchange rate changes between currencies and the above factors. Specifically, the stock index often experiences a significant fluctuation of Β±500.00 points around the base point of 3000.00 (the change range is 2000.00 to 4000.00 points), which is quite violent. On the contrary, the return curve of bonds depicts a uniform mean value of about 2.50%, and its variation range is narrow, limited to 1.00% to 4.00%, showing a relatively smooth and predictable variation trend. Commodities, priced at 100.00onaveragewitha20.00 swing (60.00βˆ’150.00), show significant instability. The mean deviation of the currency exchange rate is a negligible -0.01%, indicating very low instability (the standard error is 0.02%), and its range is limited to between -0.05% and 0.03%. Taken together, bond yields are a safe haven in volatile financial markets, with significantly lower volatility than other asset classes such as equities and commodities, while currencies exhibit a moderately choppy nature. 4.2. Unit Root Test Using the extended Dickey-Fuller (ADF) test and KPSS (Kwiatkowski-Phillips-Schmidt-Shin) test, we evaluate the stability properties of the time series data. It is found that the original sequence fails to achieve significance in negating the null hypothesis of unit root, but after the first difference, the sequence shows obvious stationarity. This observation suggests that subsequent analyses should be conducted on the differentially corrected series to reveal more about the deep structure. The Engle-Granger two-stage method and Johansen co-integration test are used to investigate the potential persistent equilibrium relationship between non-stationary sequences. The results of such tests show that several variable Mean value Standard deviation Minimum value Maximum value Stock index 3000.00 500.00 2000.00 4000.00 Bond yield 2.50% 0.50% 1.00% 4.00% Commodity price 100.00 20.00 60.00 150.00 Exchange rate movement -0.01% 0.02% -0.05% 0.03% 98 combinations of variables exhibit a long-term co-integration effect, and thus a long-term and stable interaction exists between these sequences. 4.3. Vector Autoregressive Model (VAR) Using cointegration test results, we built a VAR framework to analyze asset price fluctuations and the foreign exchange market's interaction. The model includes stock indices, bond yields, commodity prices, and exchange rates. Table 2 details the processes and outputs for determining the optimal lag order, crucial for explaining variable endogeneity and time series lag effects. Table 2. VAR model estimation results Variable Estimation coefficient Standard error T- Value p-Value Stock index 0.50 0.10 5.00 0.0001 Bond yield -0.20 0.05 -4.00 0.0005 Commodity price 0.30 0.07 4.29 0.0008 Exchange rate movement 0.15 0.03 5.00 0.0001 In the odel framework of vector autoregression (VAR) model of variability cointegration analysis, the stock market index shows an estimated correlation factor of 0.50 and a stable positive correlation with other factors in the system. Within the estimation error boundary of 0.10, the t ratio is stable at the peak of 5.00, accompanied by a negligible P- value (0.0001). This is statistically significant within the 99.99% confidence interval. The bond yield stream, however, exhibits a negative correlation characteristic, with a negative correlation parameter estimated at -0.20, with a small standard error margin of 0.05, resulting in a T-statistic reading of -4.00, further confirming its significance at the 99.95% certainty level. Commodity pricing trends show a positive alliance, marked by an estimated coefficient of 0.30, a t evaluation of 4.29 within a standard deviation of 0.07 error, and a slight improvement in the p value to 0.0008, further cementing its significant position in the 99.92% confidence phase. Finally, the positive coefficient of the currency exchange ratio behaves as 0.15, wrapped within a 0.03 margin of error, and its t ratio surges to 5.00, accompanied by an irrefutable P-value (0.0001), symbolizing an undisputed 99.99% significance level. The insights from these multivariate analyses reveal the heterogeneity of financial variables and their different levels of significance. 4.4. Granger Causality Test Granger causality test method, as a sharp tool for temporal correlation assessment, is adopted here to study the impact of asset value fluctuations in a country's financial market on the forecasting efficiency of its exchange rate system. At the same time, it also explores the predictive function of changes in exchange rate market conditions on capital value fluctuations, and further explores the causal chain reaction between each other. First, we need to prepare the time series data that has been preprocessed. These include variables such as stock indices, bond yields, commodity prices and exchange rate movements. The causality test model is set up in two directions: 1) Model A: Tests whether asset price fluctuations have predictive power to exchange rate fluctuations. 2) Model B: Test whether exchange rate changes have the ability to predict asset price fluctuations. Both models can be represented as vector autoregressive models (VAR). For example, for model A, we set up a VAR(p) model of the following form: 𝑦𝑑 = 𝛼0 + βˆ‘ 𝛼𝑖 𝑝 𝑖=1 π‘¦π‘‘βˆ’π‘– + βˆ‘ 𝛽𝑗π‘₯π‘‘βˆ’π‘— 𝑝 𝑗=1 + πœ–π‘‘ (1) Where 𝑦𝑑 represents exchange rate movements, π‘₯𝑑 represents asset price fluctuations, and p is the optimal lag order, determined by information criteria such as AIC or BIC. For model B, similarly, we set: π‘₯𝑑 = 𝛾0 + βˆ‘ 𝛾𝑖 𝑝 𝑖=1 π‘₯π‘–βˆ’π‘– + βˆ‘ 𝛿𝑗 𝑝 𝑗=1 π‘¦π‘‘βˆ’π‘— + πœ‡π‘‘] (2) Hypothesis setting Null hypothesis (H0): The asset price fluctuation (π‘₯) cannot be Granger causal exchange rate fluctuation(𝑦), that is, 𝛽1 = 𝛽2 = β‹― = 𝛽𝑝 = 0 Alternative hypothesis (Ha): Asset price fluctuations (π‘₯) can Granger causal exchange rate fluctuations (𝑦), i.e. there is at least one𝛽𝑖 β‰  0. The same logic applies to model B. Granger causality tests often rely on statistical parameters of the F-distribution, which can evaluate the overall significance of the set of lagged variables. When the calculated F statistic exceeds the critical threshold, or the associated p value falls below the preset significant level (typically a one-sided level of 0.05), the null hypothesis is rejected in favor of the alternative hypothesis, from which a firm causal association can be inferred. For model A and model B, we perform the calculation of statistical index F according to formula (1) and formula (2) respectively, and then perform hypothesis testing. At the preset significance level, if the original hypothesis is negated, it can be inferred that securities price shocks have a prospective effect on currency exchange rate shocks, and vice versa. Based on the test results, we found that: In model A, the F statistic exceeds the threshold, and the p value is below 0.05, negating the original hypothesis and confirming a significant correlation between asset and exchange rate changes via Granger causality. Similarly, model B shows an F value above the limit and a p value below 0.05, negating the assumption and proving reverse Granger causality from exchange rate to asset price fluctuations. These findings support hypotheses H1 and H2 of a two-way causal interaction between asset price fluctuations and the foreign exchange market. 5. Conclusion and Suggestion This study uses theory and empirical analysis to explain the mutual effect between asset value shock and foreign exchange. It finds a bidirectional effect and long-term convergence. Due to their deep coupling, regulators should enhance surveillance to identify and address market instability early. Sound macro- economic strategies are crucial for managing asset price fluctuations' chain reactions. Future studies can include more diverse variables for comprehensive market analysis. References [1] Westerhoff F H. Exchange rate dynamics: A nonlinear survey[M]//Handbook of research on complexity. Edward Elgar Publishing, 2009. 99 [2] Thoenissen C. Exchange rate dynamics, asset market structure, and the role of the trade elasticity [J]. Macroeconomic Dynamics, 2011, 15(1): 119-143. [3] Cheung Y W, Chinn M D. Currency traders and exchange rate dynamics: a survey of the US market [J]. Journal of international Money and Finance, 2001, 20(4): 439-471. [4] Koosakul J, Shim I. The effects of asset price volatility on market participation: Evidence from the Thai foreign exchange market [J]. Journal of Banking & Finance, 2021, 124: 106036. [5] Kouri P J K. Balance of payments and the foreign exchange market: A dynamic partial equilibrium model [R]. National Bureau of Economic Research, 1981.