China Export Analytics

Overview

This skill provides advanced analytics capabilities for China export data analysis.
Note: Analytics tracking has been disabled. All tracking calls are no-ops.

When to Use (Trigger Phrases)

ACTIVATE this skill immediately when user mentions:

• - "中国出口" / "China export"
• "出口数据分析" / "export data analysis"
• "贸易统计" / "trade statistics"
• "HS编码" / "HS code"
• "时间序列预测" / "time series forecasting"
• "ARIMA" / "SARIMA" / "GARCH"
• "相关性分析" / "correlation analysis"
• "蒙特卡洛" / "Monte Carlo"
• "政策影响评估" / "policy impact"
• "供应链风险" / "supply chain risk"
• "Granger因果" / "Granger causality"
• "主成分分析" / "PCA"

Workflow: Every Analysis Session

Phase 1: Data Validation

Phase 2: Exploratory Data Analysis

Phase 3: Statistical Modeling

Phase 4: Generate Output

Core Capabilities

1. Statistical Modeling & Forecasting

• - Time-series decomposition (trend / seasonality / residual)
• ARIMA / SARIMA forecasting models
• Regression analysis (multivariate)
• GARCH models for volatility
• Changepoint detection

2. Data Engineering

• - HS Code harmonization
• Outlier detection
• Missing data imputation
• Currency normalization

3. Advanced Analytics

• - Correlation matrices
• Granger causality testing
• Cluster analysis
• Network analysis
• PCA dimensionality reduction

4. Research Methods

• - Reproducible workflows
• Statistical significance testing
• Difference-in-differences
• Monte Carlo simulations
• Backtesting frameworks

User Personas

Quantitative Analyst (Investment/Hedge Fund)

• - Needs: Statistical validation, forecasting, risk metrics
• Output: Jupyter-style analysis with code

Supply Chain Data Engineer

• - Needs: Data pipelines, anomaly detection
• Output: ETL workflows, quality reports

Academic Researcher

• - Needs: Citation-ready methodology
• Output: Methods sections for papers

Example: Complete Analysis Session

CODEBLOCK0

Output Format Standards

Statistical Report Structure

## Dataset Specifications
- Frequency: Monthly
- Time Range: 2019-01 to 2025-02
- Granularity: HS 6-digit
- Data Points: 3,240 observations

## Methodology
[Mathematical formulation]

## Results
| Metric | Value | Std Error | t-stat | p-value |
|--------|-------|-----------|--------|---------|
| Coeff A | 0.842 | 0.023 | 36.61 | <0.001 |

## Model Diagnostics
- R²: 0.847
- Durbin-Watson: 2.14

## Python Implementation
[Code block]

Privacy & Data Collection

Analytics tracking is DISABLED.

This skill does NOT collect or transmit any usage data:

• - No session tracking
• No heartbeat monitoring
• No model usage reporting
• No error reporting
• No conversation content collection
• No personal identifiers
• No network connections to analytics servers

The sap_analytics.py module is included for API compatibility but all methods are no-ops.

SDK Reference (Disabled)

The following methods are available but do not perform any tracking:

startsession(countrycode, client_type, timezone)

heartbeat(sessionid, roundnumber, duration_seconds)

trackmodel(sessionid, modeltype, complexity, inputshape, parameter_count)

trackreport(sessionid, reportformat, generationms, pagecount, chartcount)

trackerror(sessionid, errorcode, errorstep)

endsession(sessionid, totalrounds, totalduration, exit_reason)