I'm seeking guidance on developing a robust quantitative model to predict silver prices using macroeconomic indicators. How can I incorporate variables like GDP growth, inflation rates, and monetary policy changes into a predictive model for silver? Which statistical techniques and models (e.g., ARIMA, VAR, machine learning) have shown success in accurately forecasting precious metal prices in response to macroeconomic shifts? Recommendations on datasets and model validation methods would also be helpful.
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
Developing a robust quantitative model to predict silver prices using macroeconomic indicators requires a careful consideration of the relationships between these indicators and the price of silver. Ideally, you should understand theoretical relationship between silver price and macro variables. In other words, what macro variables affect silver price, what is lag-structure between variables, is the functional form linear or non-linear, among other questions. If "theory" is available, you can empirically estimate the model (estimate parameters of the model).
However, most of the time theoretical relationship is not available. Therefore, you should play around with empirical data and find best model having minimum test error. You can use both econometric (e.g., ARIMAX, or multivariate regression models) or machine learning models.
Have a look at the following paper: Gargano, A., & Timmermann, A. (2014). Forecasting commodity price indexes using macroeconomic and financial predictors. International Journal of Forecasting, 30(3), 825-843. (https://www.sciencedirect.com/science/article/abs/pii/S016920701300099X)
Other papers that might be relevant:
- Wang, J., & Li, X. (2018). A combined neural network model for commodity price forecasting with SSA. Soft Computing, 22, 5323-5333.
- Kohzadi, N., Boyd, M. S., Kermanshahi, B., & Kaastra, I. (1996). A comparison of artificial neural network and time series models for forecasting commodity prices. Neurocomputing, 10(2), 169-181.
