Description
This research investigates the relationship between daily price indicators and trading volume for the SPDR S&P 500 ETF Trust (SPY), one of the most actively traded exchange-traded funds (ETFs) in global markets. The dataset includes historical daily data for Open, High, Low, and Close prices, trading Volume, Dividends, and corporate actions, making it an extensive resource for financial modeling and empirical analysis. A multiple linear regression model was applied using 251 daily observations, with trading volume as the dependent variable and price variables as independent predictors. The model demonstrated a moderately strong explanatory power, with a multiple correlation coefficient of R=0.758 and an adjusted R2=0.567, indicating that approximately 56.7% of the variance in volume is explained by the predictors. The model was statistically significant overall (F = 82.886, p < .001). Individually, the High price (t = 7.83, p < .001) and Low price (t = -11.36, p < .001) were significant predictors of trading volume, while Open (p = .154) and Close (p = .191) prices were not statistically significant. Pearson correlation analysis revealed very strong multicollinearity among the independent variables (correlation coefficients exceeding 0.98), suggesting redundancy that may compromise the interpretability of individual coefficients in the regression model. These results highlight the relevance of intraday price extremes (High and Low) in forecasting trading activity and emphasize the need for multicollinearity control in financial regression modeling. The findings provide a practical foundation for quantitative analysts and financial researchers developing predictive models or algorithmic trading strategies based on price-volume relationships.
| Affiliation / University / Organization | Universidad Ana G. Méndez, Recinto de Gurabo |
|---|
