GOLD/SILVER RATIO: GenAI with Quant Agents on Microsoft AI Foundry | R bloggers

1. Introduction: Agentic Finance’s strategic edge

In today’s quantitative finance landscape, the bottleneck is no longer the availability of data, but the speed at which insights are generated. Taking advantage of the Microsoft AI Foundry ecosystem we have moved beyond static scripting into the domain of Autonomous financial agents. This article explores how a specialist agent can manage precious metals volatility by analyzing the gold/silver ratio with high precision.

2. Infrastructure: Model deployment on Microsoft AI Foundry

The intelligence behind this analysis is not a local script, but a deployed model instance on Microsoft AI Foundry. We use the GPT-4o model, deployed as a scalable web service within the Foundry environment.

• Endpoint security: By using the Azure OpenAI service within AI Foundry, we ensure that financial queries and data remain within a secure, enterprise-level environment.
• Agentic logic: The “Agent” is more than just a model; it is a programmed entity with a specific character System prompt that defines his persona as a quantitative researcher. This allows the model to ‘reason’ through the necessary steps: from loading the library to merging data and final visualization.

3. The Technical Bridge: Python-R Integration

One of the most powerful features of our AI Foundry Agent is its multi-language capability. It bridges the gap between Python and R using the rpy2 library, creating a high-quality research pipeline.

The R ecosystem in play:

• tidyquant & timetk: These packages drive our time series analysis. tidyquant allows for seamless retrieval of GC=F And SI=F data, while timetk manages the complex task of plotting with built-in smoothing algorithms.
• dplyr & lubridate: Essential for “cleanly” manipulating data, allowing the Agent to perform inner_join operations and date-based filtering with surgical precision.

4. Methodology: tame the noise with visual precision

To extract useful trends, the agent is instructed to create a LOESS smoothing algorithm. By being strict .line_size = 1.5 And .smooth_size = 1.5we ensure that the trendline is bold enough to be the primary focus for analysts, effectively ‘taming’ daily price volatility.

5. Conclusion: Scaling up quantitative research

The synergy between Microsoft AI Foundry, LLMs deployedand specialized R packages represents the future of financial research. We replaced manual data management with an autonomous, standardized agent that can scale across thousands of different asset pairs with a single command.

The ABI connection (Bridging Python to R in VS Code)

To run the script locally US codewe need to set up a robust system Binary Application Interface (ABI) connection. This is provided by the rpy2 library, which serves as a translation layer between Python and the R interpreter.

• Synchronization: The script uses a localconverter to convert Python data types into R objects in real time.
• Environment Sync: Before executing the agent’s code, the script automatically syncs the workbook (setwd) so that files generated by R (such as the ratio_plot.png) are immediately accessible to the Python environment for rendering.

import os

# Force rpy2 to use ABI mode to avoid the Windows CFFI conflict
os.environ['RPY2_CFFI_MODE'] = 'ABI'

import rpy2.robjects as robjects
from rpy2.robjects.packages import importr
print("Interface initialized in ABI mode.")

The integrated agent script:

import os
import httpx
from openai import AzureOpenAI
import rpy2.robjects as robjects
from rpy2.robjects import pandas2ri
from rpy2.robjects.conversion import localconverter
from IPython.display import Image, display

#Microsoft AI Foundry - Azure OpenAI Connection
client = AzureOpenAI(
    api_version="2024-12-01-preview",
    azure_endpoint="AZURE_OPENAI_ENDPOINT",
    api_key="AZURE_OPENAI_KEY", 
    http_client=httpx.Client(verify=False, trust_env=False)
)

def run_updated_agent(user_request):
    system_instructions = (
        "You are a Quantitative Researcher. MANDATORY: All output, comments, and labels in English. "
        "Strict Operational Guidelines:\n"
        "1. Libraries: library(tidyquant), library(timetk), library(lubridate), library(dplyr), library(ggplot2).\n"
        "2. Analysis: Fetch GC=F and SI=F for 3 years, merge via inner_join, and calculate 'ratio'.\n"
        "3. Visualization: Use timetk::plot_time_series with .interactive = FALSE and .title = \"Gold/Silver Ratio\".\n"
        "4. Precision: Set .line_size = 2 and ALWAYS set .smooth_size = 2 for the smoothing line.\n"
        "5. Set title font face and axis texts font face to 'bold', and size to 16 with theme() function.\n"
        "6. EXPORT: Save using 'ggsave(\"ratio_plot.png\", width = 10, height = 6)'.\n"
        "7. Output ONLY raw R code."
    )

    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[
            {"role": "system", "content": system_instructions},
            {"role": "user", "content": user_request}
        ]
    )
    
    # Cleaning any markdown or headers to get raw code
    agent_code = response.choices[0].message.content.strip()
    if agent_code.startswith("```"):
        agent_code = "\n".join(agent_code.split("\n")[1:-1])
    
    print("-" * 40)
    print(agent_code)
    print("-" * 40)
    
    try:
        with localconverter(robjects.default_converter + pandas2ri.converter):
            # Synchronize working directory
            robjects.r(f'setwd("{os.getcwd().replace("\\", "/")}")')
            robjects.r(agent_code)
            
            if os.path.exists("ratio_plot.png"):
                display(Image(filename="ratio_plot.png"))
    except Exception as e:
        print(f"Agent Error: {e}")

# Execution
run_updated_agent("Plot the Gold/Silver ratio for the last 3 years with a smooth line.")

Related