Predicting the Price of Gold in the Financial Markets Using Hybrid Models

Abstract: Predicting the price that has the least error and can provide the best and highest accuracy has been one of the most challenging issues and one of the most critical concerns among capital market activists and researchers. Therefore, a model that can solve problems and provide results with high accuracy is one of the topics of interest among researchers. In this project, using time series prediction models such as ARIMA to estimate the price, variables, and indicators related to technical analysis show the behavior of traders involved in involving psychological factors for the model. By linking all of these variables to stepwise regression, we identify the best variables influencing the prediction of the variable. Finally, we enter the selected variables as inputs to the artificial neural network. In other words, we want to call this whole prediction process the "ARIMA_Stepwise Regression_Neural Network" model and try to predict the price of gold in international financial markets. This approach is expected to be able to be used to predict the types of stocks, commodities, currency pairs, financial market indicators, and other items used in local and international financial markets. Moreover, a comparison between the results of this method and time series methods is also expressed. Finally, based on the results, it can be seen that the resulting hybrid model has the highest accuracy compared to the time series method, regression, and stepwise regression.

An Analytical Critique of Hybrid Models in Gold Price Prediction

The paper titled "Predicting the Price of Gold in the Financial Markets Using Hybrid Models" explores the development and application of a hybrid forecasting model designed to predict gold prices in international financial markets with enhanced accuracy. The hybrid model discussed in this study amalgamates traditional econometric and time-series methods with technical analysis and machine learning techniques, notably artificial neural networks (ANNs). This fusion aims to leverage the strengths of each individual method to minimize prediction errors associated with gold prices.

Methodology Overview

The authors implement a multi-step hybrid model, identified as the "ARIMA - Stepwise Regression - Neural Network" approach.

1. ARIMA Time Series Models: This foundational component utilizes the autoregressive integrated moving average (ARIMA) method to handle non-stationary data through differencing, assessing autocorrelation, and fitting a model conducive to capturing temporal dependencies within financial time series data. The ARIMA model's role primarily focuses on predicting future values based on past behaviors, enabling baseline forecasting of gold prices.

2. Technical Analysis Indicators: A suite of five technical indicators is deployed—Exponential Moving Average (EMA), Relative Strength Index (RSI), Stochastic Oscillator (\%K and \%D), and Williams \%R—to capture trading patterns and investor sentiment. These indicators offer quantitative measures of price trends and trader behaviors, functioning as essential inputs for subsequent predictive algorithms.

3. Stepwise Regression: The paper transitions into a statistical refinement phase, employing stepwise regression to discern and select the most relevant predictive technical indicators and econometric parameters. This selection process aids in reducing model complexity and avoiding overfitting by retaining only the most statistically significant variables.

4. Artificial Neural Network (ANN): Finally, the ANN synthesizes information from selected inputs to predict future gold prices. The ANN benefits from the streamlined inputs derived from stepwise regression, enhancing predictive accuracy and operational speed. The network exhibited superior prediction accuracy, as corroborated by an impressive MAPE metric value.

Key Results

The hybrid model showcases commendable performance metrics, with the ANN stage yielding a prediction accuracy of 99.29%, notably outperforming isolated ARIMA models and traditional regression forecasts. Such results underscore the efficacy of hybrid models in financial forecasting, particularly the strategic integration of AI methodologies to boost prediction reliability.

Implications and Future Directions

The implications of this research are multi-fold. Practically, it signifies a robust predictive tool capable of aiding financial analysts, traders, and institutional investors in pre-emptively gauging market movements. Theoretically, it expands the horizon for employing hybrid models in forecasting not only gold prices but also other commodities, stock indices, and market indicators with minimal adjustment.

Future pursuits could involve incorporating alternative variable selection techniques such as Random Forests or Genetic Algorithms to further optimize model inputs. Additionally, infusing fundamental variables could enrich forecasts with macroeconomic insights, enhancing robustness against broader market fluctuations.

Overall, the paper presents a comprehensive exploration of leveraging hybrid models for financial market predictions, portraying significant advancements in the quest for precision in econometric modeling commingled with technological sophistication.