Watching gold closely matters to people navigating world finance, since getting its value right helps them handle unclear situations better. What follows describes a digital tool built for forecasting gold prices, pulling together smart algorithms plus up-to-the-minute money-market numbers so users see trends, decide timing around purchases or exits, explore visuals shaped by personal choices within the system’s interface. Instead of relying on one method alone, it blends fresh pricing figures gathered via Gold API with forecasts made by a trained math model fed on past patterns – opening, highest, lowest, closing values alongside trade amounts recorded each day and month pulled through yfinance from Yahoo Finance. The core idea leans on balance: merging real-time signals with calculated outlooks using adjusted importance levels across inputs inside this mixed strategy setup. One part shows a real-time view of gold prices on the main screen. Following that, price history appears through candlestick patterns showing open, high, low, close shifts. Instead of just past trends, predictions pop up using modeled outputs to suggest when trading moves might work. Each day, upcoming week estimates form by running daily updates based on repeating calculation logic. Signals also emerge from momentum analysis, specifically RSI readings guiding potential start points. Lastly, details about how everything fits together – methods, structure, flow – appear in a breakdown area explaining what runs beneath. Out of Python and Flask came the app’s engine, while its look took shape through HTML tied with CSS plus a dash of Bootstrap – all made lively with Chart.js for smoother interaction. Seven different models stepped forward, each landing near 0.94 on the R2 scale, backed by error rates sitting at 120 (RMSE) and 85 (MAE), hinting tightly at consistent precision. When tested by actual users, reactions tilted favorable, responses lighting up where it counts – especially under pressure of live money choices. So it stands: those navigating daily trades or forecasts in finance might find this quietly useful.

The suggested solution is realized via a full-fledged web-based system that involves the usage of such tools as Ridge Regression Model together with financial market APIs for acquiring live data. The main functions of the system are related to tracking the real-time dynamics of gold prices, plotting visual graphics reflecting the price dynamics, predicting the future gold prices with the help of machine learning models, making multi-day price forecasts, and suggesting the optimal trading decisions using RSI values . According to the results of the experiment conducted, the suggested solution demonstrates very high prediction quality parameters: R² = 0.94, RMSE = 119.87, and MAE = 84.63 . The proposed solution proves that a proper combination of predictions of machine learning with real-time market data leads to stable forecasting results. As the prediction model applies the method of autoregressive forecasting with smoothing, the forecast for the next seven days becomes consistent and realistic. Moreover, according to the evaluation conducted with real users, the solution proves effective from a practical perspective, achieving a high level of satisfaction in a number of aspects. For example, users gave their highest ratings regarding the quality of trading signals – an average rating of 4.5 out of 5 – and were satisfied with the visualization of results – an average rating of 4.3 out of 5 . First, in the future, the model can be enhanced even further through the application of advanced data modeling techniques such as applying machine learning algorithms like long-short term memory (LSTM) and transformer. This is because the machine learning algorithms will help in modeling time series data including the gold price variations over time. Secondly, the model should incorporate other economic indicators such as the DXY index, CPI, and the TIPS spreads in order to enhance the ability of the system to identify the factors that affect the gold price fluctuations. In addition, the implementation of natural language processing (NLP) can provide more useful information regarding the financial market by analyzing financial news. Thirdly, implementing portfolio optimization using the gold price predictions is one interesting aspect of this project that may be considered in the future.

[1]C. J. Ntim, K. A. Opong and J. Danbolt, “The relative value relevance of shareholder versus stakeholder corporate governance disclosure policy reforms in South Africa,” Corporate Governance: An International Review, vol. 20, no. 1, pp. 84–105, 2012. [2]R. F. Engle, “Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation,” Econometrica, vol. 50, no. 4, pp. 987– 1007, 1982. [3]J. Patel, S. Shah, P. Thakkar and K. Kotecha, “Predicting stock and stock price index movement using trend deterministic data preparation and machine learning techniques,” Expert Systems with Applications, vol. 42, no. 1, pp. 259–268, 2015. [4]A. E. Hoerl and R. W. Kennard, “Ridge regression: biased estimation for nonorthogonal problems,” Technometrics, vol. 12, no. 1, pp. 55–67, 1970. [5]X. Li, P. Li and T. Wang, “Gold price forecasting using LSTM neural networks,” in Proceedings of IEEE DSIT, Guangzhou, China, 2020, pp. 1–6. [6]S. Makridakis, E. Spiliotis and V. Assimakopoulos, “The M4 competition: 100,000 time series and 61 forecasting methods,” International Journal of Forecasting, vol. 36, no. 1, pp. 54–74, 2020. [7]J. W. Wilder, New Concepts in Technical Trading Systems. Greensboro, NC: Trend Research, 1978. [8]M. C. Kung and B. T. Carverhill, “When is the short rate drift long run mean reverting?” Review of Finance, vol. 9, no. 3, pp. 411–429, 2005. [9]G. Biau and E. Scornet, “A random forest guided tour,” TEST, vol. 25, no. 2, pp. 197–227, 2016. [10]R. Aroussi, “yfinance: Download market data from Yahoo Finance API,” GitHub repository, 2019. Available: https://github.com/ranaroussi/yfinance [11]F. Pedregosa et al., “Scikit-learn: machine learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011. [12]S. Kumar and R. Sharma, “Hybrid machine learning models for gold price prediction using technical indicators,” International Journal of Engineering Research and Technology, vol. 13, no. 3, pp. 412–420, 2020. [13]Y. Chen, Z. Zheng and F. Li, “Comparative study of machine learning algorithms for financial time series prediction,” in Proceedings of IEEE BDAI, 2021, pp. 78–84. [14]M. Ballings, D. Van den Poel, N. Hespeels and R. Gryp, “Evaluating multiple classifiers for stock price direction prediction,” Expert Systems with Applications, vol. 42, no. 20, pp. 7046–7056, 2015. [15]N. Jain and A. Kumar, “Real-time financial data integration in web-based prediction systems using REST APIs,” International Journal of Advanced Computer Science and Applications, vol. 12, no. 7, pp. 345–352, 2021. [16]P. Sharma and V. Gupta, “Flask-based microservices architecture for machine learning model deployment in financial applications,” in Proceedings of ICCCIS, 2022, pp. 102–108. [17]A. Banerjee, R. Ghosh and S. Das, “Ridge regression for financial feature sets with high multicollinearity: a comparative analysis,” Journal of Statistical Computation and Simulation, vol. 91, no. 5, pp. 1012–1028, 2021. [18]K. Guo, Y. Zhou and X. Shi, “An empirical study of RSI-based trading strategies on commodity futures markets,” Journal of Futures Markets, vol. 40, no. 6, pp. 917–935, 2020. [19]J. B. Heaton, N. G. Polson and J. H. Witte, “Deep learning for finance: deep portfolios,” Applied Stochastic Models in Business and Industry, vol. 33, no. 1, pp. 3–12, 2017. [20]A. Yadav and P. Singh, “A survey on web-based financial analytics platforms leveraging machine learning,” International Journal of Information Technology, vol. 14, no. 4, pp. 1893–1902, 2022.

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close