var
GitHub用于分析多个同步时间序列的动态关系。通过验证数据对齐与平稳性,执行向量自回归(VAR)计算,输出滞后选择、预测表现及动态响应结果,并生成可视化图表,将格兰杰关系表述为预测信息而非因果证明。
Trigger Scenarios
需要分析多个时间序列间的动态影响
进行向量自回归(VAR)建模
评估时间序列的格兰杰因果关系
Install
npx skills add Zafer-Liu/Data-Analysis-Agent --skill var -g -y
SKILL.md
Frontmatter
{
"icon": "🔗",
"name": "var",
"description": "使用 VAR 分析多个时间序列的动态关系(forecast 向量自回归)",
"allowedTools": [
"get_schema",
"query_data",
"run_analysis",
"generate_chart"
]
}
VAR 分析
确认多个同步时间序列、频率和滞后范围,检查平稳性与样本长度。报告滞后选择、预测表现和动态响应;将格兰杰关系表述为预测信息而非因果证明。
Tool routing
- Use
get_schemato identify multiple time-aligned target series, the time column, and source table. - Use
query_dataonly to verify frequency alignment, missing periods, stationarity preparation needs, and sample length. - Use
run_analysiswithanalysis_name="Time_Series_VAR"for the actual VAR computation. - Use
generate_charton forecast, lag, or response result tables afterrun_analysissucceeds.
Implementation reference
- Tool entry:
agent/tools/business/data.py::_tool_run_analysis - Analysis registry:
Function/Analyze/registry.py - Analysis implementation:
Function/Analyze/Time_Series_VAR/analyze.py - Chart implementation:
Function/Charts_generation/chart_generate.py
Version History
- d6a2c3e Current 2026-07-24 12:13


