asset-monitor

To detect daily asset price anomalies using z-score, this Skill compares recent returns to historical mean and standard deviation. You can run python scripts/monitor.py to flag moves exceeding configurable thresholds across sector indices, stocks, bonds, ETFs, and commodities.

Yes, you need a running Wind terminal to run anomaly detection scripts. The Skill requires WindPy and access to the windpy-sdk data model, alongside pandas and numpy, to fetch historical data and calculate z-scores for financial analysis.

You can screen daily return outliers across multiple asset classes. Supported classes include major equity indices, sector indices, bonds, ETFs, commodities, and global indices, allowing comprehensive financial risk-management monitoring.

Yes, you can generate Excel and Markdown reports for risk anomalies. The Skill automates financial reporting by outputting flagged z-score outliers into these formats, supporting scheduled runs and sharing risk insights with your team.

Z-score based anomaly detection works by measuring how many standard deviations a recent return is from its historical mean. This surfaces meaningful outliers in financial analysis when moves exceed your configurable threshold limits.

A limitation of using z-score for asset monitoring is that it assumes historical mean and standard deviation are reliable baselines. Sudden regime shifts or volatility clustering may cause false positives, requiring threshold adjustments for different asset classes.