Portfolio Optimizer
✓ Official Vendor SpecFinanceBankingapiKey158 EndpointsREST
For Agents
Optimize investment portfolios using mean-variance analysis, calculate Value-at-Risk, estimate covariance matrices, run Monte Carlo simulations, and compute risk-adjusted returns across up to 100 assets.
Use for: I need to find the minimum variance portfolio for my assets, Calculate the Value-at-Risk of my portfolio, I want to optimize for maximum Sharpe ratio, Estimate the correlation matrix for a set of asset returns
Not supported: Does not handle trade execution, market data sourcing, or accounting — use for quantitative portfolio analysis and optimization only.
Portfolio Optimizer is a quantitative finance API offering 158 endpoints for analyzing and optimizing investment portfolios. It covers modern portfolio theory (mean-variance optimization, maximum Sharpe ratio, minimum variance), risk analysis (Value-at-Risk, Conditional VaR), asset correlation and covariance matrix estimation, hierarchical clustering, Monte Carlo simulation, factor models (Fama-French), and capital market assumptions. The API supports up to 100 assets and 2500 data points per authenticated request.
Install Jentic One Beta
Connect the Portfolio Optimizer to your agent
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Portfolio Optimizer, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.
Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.
1
Step 1: Jentic One Host machine
# On the machine that will host your Jentic One instance:
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh2
Step 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh
jentic register # connects your agent to your Jentic One instanceJentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.
Capabilities
What an agent can do with Portfolio Optimizer API.
Optimize portfolios for minimum variance, maximum Sharpe ratio, or target return on the efficient frontier
Calculate empirical, Gaussian, and Cornish-Fisher Value-at-Risk and Conditional VaR
Estimate asset covariance and correlation matrices with shrinkage and denoising methods
Run Monte Carlo return simulations with bootstrap and Gaussian models
Apply hierarchical clustering and FTCA for asset grouping and diversification analysis
Retrieve Fama-French factor model data for US, developed, and emerging markets
Compute capital market assumptions averaged from leading financial institutions
Use Cases
Patterns agents use Portfolio Optimizer API for, with concrete tasks.
★ Mean-Variance Portfolio Optimization
Compute optimal asset weights using modern portfolio theory. The API provides minimum variance, maximum return, maximum Sharpe ratio, and mean-variance efficient portfolio optimizations. Each supports diversified variants, resampling-based methods for robustness, and subset resampling for large universes. Input asset returns and constraints, receive optimal weights.
Submit 5 asset return series to the /portfolio/optimization/maximum-sharpe-ratio endpoint and retrieve the optimal weight allocation
Portfolio Risk Measurement
Quantify portfolio downside risk using multiple Value-at-Risk methodologies. The API offers empirical, Gaussian, Cornish-Fisher (with corrections), and historical VaR calculations, plus Conditional VaR variants. Each accepts portfolio weights and return data, returning risk metrics at configurable confidence levels.
Calculate the 95% Cornish-Fisher Value-at-Risk for a portfolio of 3 assets using /portfolio/analysis/value-at-risk/cornish-fisher with their historical returns
Covariance Matrix Estimation
Estimate robust covariance and correlation matrices from asset return data using standard sample estimation, shrinkage methods, exponential weighting, and denoising techniques. The API supports up to 100 assets and provides both historical and forecast covariance matrices for use in downstream optimization.
Submit 10 asset return series to the covariance matrix estimation endpoint with Ledoit-Wolf shrinkage and retrieve the estimated matrix
Monte Carlo Return Simulation
Generate simulated return scenarios for stress testing and scenario analysis. The API offers bootstrap simulations (sampling from historical returns) and Gaussian simulations (parametric generation from mean and covariance). Results feed into optimization and risk analysis endpoints for comprehensive portfolio evaluation.
Run a Gaussian Monte Carlo simulation of 1000 scenarios for 4 assets using /assets/returns/simulation/monte-carlo/gaussian with their mean returns and covariance matrix
AI Agent Quantitative Analysis via Jentic
AI agents use Jentic to discover Portfolio Optimizer operations by intent, such as 'optimize portfolio for maximum Sharpe ratio' or 'calculate Value-at-Risk'. Jentic returns the operation schema with required input formats (arrays of asset returns, weight vectors), then handles X-API-Key injection for authenticated access to higher rate limits.
Search Jentic for 'optimize portfolio for minimum variance', load the operation schema, and execute with 5 asset return series to get optimal weights
Key Endpoints
158 endpoints — portfolio optimizer is a quantitative finance api offering 158 endpoints for analyzing and optimizing investment portfolios.
METHOD
PATH
DESCRIPTION
POST
/portfolio/optimization/maximum-sharpe-ratio
Find the portfolio with the highest risk-adjusted return
POST
/portfolio/optimization/minimum-variance
Compute the minimum risk portfolio
POST
/portfolio/analysis/value-at-risk/cornish-fisher
Calculate Cornish-Fisher VaR
POST
/portfolio/optimization/mean-variance-efficient
Find an efficient frontier portfolio for a target return
POST
/assets/returns/simulation/monte-carlo/gaussian
Run Gaussian Monte Carlo return simulations
POST
/assets/clustering/hierarchical
Hierarchical clustering of assets
GET
/factors/models/fama-french/us
Retrieve U.S. Fama-French factor data
GET
/markets/capital-assumptions/usd
Get USD capital market assumptions
POST
/portfolio/optimization/maximum-sharpe-ratio
Find the portfolio with the highest risk-adjusted return
POST
/portfolio/optimization/minimum-variance
Compute the minimum risk portfolio
POST
/portfolio/analysis/value-at-risk/cornish-fisher
Calculate Cornish-Fisher VaR
POST
/portfolio/optimization/mean-variance-efficient
Find an efficient frontier portfolio for a target return
POST
/assets/returns/simulation/monte-carlo/gaussian
Run Gaussian Monte Carlo return simulations
POST
/assets/clustering/hierarchical
Hierarchical clustering of assets
GET
/factors/models/fama-french/us
Retrieve U.S. Fama-French factor data
GET
/markets/capital-assumptions/usd
Get USD capital market assumptions
Why Jentic?
Three things that make agents converge on Jentic-routed access.
Credential isolation
Portfolio Optimizer X-API-Key credentials are stored encrypted in the Jentic vault. Agents receive scoped access to authenticated endpoints with higher rate limits, without the API key appearing in agent context.
Intent-based discovery
Agents search by intent (e.g., 'optimize portfolio for maximum Sharpe ratio' or 'calculate VaR') and Jentic returns the matching operation from 158 available endpoints with the required input schema for asset returns and constraints.
Time to first call
Direct Portfolio Optimizer integration: 2-4 days for understanding 158 endpoints, formatting asset return arrays, and handling rate limits. Through Jentic: under 1 hour — search, load schema, execute.
Related APIs
Alternatives and complements available in the Jentic catalogue.
FAQs
Specific to using Portfolio Optimizer API through Jentic.
What authentication does the Portfolio Optimizer API use?
The API uses an optional X-API-Key header for authenticated access. Anonymous users get limited access (1 request per second, 20 assets, 500 data points). Authenticated users get 10,000 requests per 24 hours, 100 assets, 2,500 data points, and 2.5 seconds execution time. Through Jentic, the API key is stored encrypted and injected automatically.
Can I compute the efficient frontier with the Portfolio Optimizer API?
Yes. The /portfolio/optimization/mean-variance-efficient endpoint finds a portfolio on the efficient frontier for a given target return. Use it repeatedly with different target returns to trace the full frontier. Diversified and resampling-based variants improve robustness for real-world asset data with estimation error.
What are the rate limits for the Portfolio Optimizer API?
Anonymous users are limited to 1 request per second (shared across all anonymous users), 1 second execution time, 20 assets, and 500 data points. Authenticated users with an API key get 10,000 requests per 24 hours with concurrent access, 2.5 seconds execution time, 100 assets, and 2,500 data points. Rate limit details are returned in x-ratelimit-* response headers.
How do I calculate Value-at-Risk through the Portfolio Optimizer API with Jentic?
Install the SDK with pip install jentic, then search for 'calculate portfolio Value-at-Risk'. Jentic returns the VaR operation schema (empirical, Gaussian, or Cornish-Fisher variants). Submit portfolio weights and historical asset returns to receive the VaR estimate at your chosen confidence level.
Is the Portfolio Optimizer API free to use?
Yes, with limitations. Anonymous access is free but restricted to a subset of endpoints, 20 assets, and 1 request per second. Full access requires an API key obtained through a small monthly contribution. Contact portfoliooptimizer.io for pricing details.
What optimization methods are available beyond mean-variance?
The API offers hierarchical risk parity, inverse volatility weighting, equal risk contribution, maximum diversification, and multiple risk-based optimization objectives including minimum tracking error and maximum Ulcer ratio. It also supports resampling-based variants of each method for improved out-of-sample robustness.
GET STARTED