Vedrai Essentials report are produced using the Vedrai Spa proprietary P.A.F. technology (Prescriptive Analytics Framework), alongside the application of generative artificial intelligence algorithms and analytical approaches.
Data Collection:
The initial phase involves gathering commodities-related data from top-tier worldwide data providers. This includes historical data, news, and other vital indicators.
The most relevant methods used for analysis purpose are:
Seasonal Analysis of Time Series Seasonal analysis techniques are applied to identify patterns and trends in commodity price time series, aiding in the understanding of temporal variability.
Text Mining and Sentiment Analysis Text mining algorithms are employed to extract information from texts related to commodities, such as news and reports. Sentiment Analysis quantifies the positive or negative sentiment associated with these sources and can generate quantitative data such as coefficients and indices.
Correlation Matrix (Pearson) A correlation matrix, using the Pearson correlation coefficient, is constructed to evaluate the relationships between variables related to commodity prices and contextual factors, such as macroeconomic variables or news. This reveals the impacts and interconnections between different variables.
Forecasting Algorithms Advanced forecasting algorithms are used, combining regression models and AI algorithms. These models take historical variables and contextual factors into account to predict future commodity prices. Decision Trees are employed to capture the complex effects that can influence future prices.
Generation of Textual Explanations Generative Artificial Intelligence is employed to generate textual explanations that accompany charts and data. These explanations provide context and interpretation of the results, facilitating the understanding of the report.