robustness_report
Check how stable a key estimate is under specification changes—dropping or adding controls, trimming, winsorizing, and subsampling—to identify violations and next steps.
Instructions
Run an automated battery of robustness checks.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"data_path": "<data_path>",
"formula": "<formula>",
"x": "<x>"
},
"name": "robustness_report"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| x | Yes | Key explanatory variable whose estimate stability is assessed. | |
| detail | No | Payload depth: 'minimal' (~150 tokens) for sub-step calls where only the point estimate is needed; 'standard' (~1K tokens) for diagnostics + coefficient table; 'agent' (~2K tokens, default) adds violations / next_steps / suggested_functions so the LLM can plan its next call without another round-trip. | agent |
| formula | Yes | Baseline regression formula, e.g. ``"y ~ x1 + x2 + x3"``. | |
| subsets | No | Named boolean masks for subsample checks. | |
| trim_pct | No | Drop observations beyond this percentile from both tails. E.g. ``0.01`` trims top and bottom 1 %. | |
| as_handle | No | If true, cache the fitted result on the server and return result_id + result_uri alongside the JSON payload so a subsequent tools/call can chain without re-running. | |
| data_path | Yes | Absolute path or URL to a data file. Supported: .csv / .tsv / .txt (delimited), .parquet / .pq, .feather / .arrow, .xlsx / .xls, .dta (Stata), .json / .jsonl. Schemes: file://, s3://, gs://, https://. | |
| result_id | No | Optional handle to a previously-fitted result (returned by an earlier call when as_handle=true). Tools that operate on a fitted object accept this in place of re-supplying data_path + columns. | |
| cluster_var | No | Column for clustered SE check. | |
| data_columns | No | Optional column projection. Parquet/Feather/Stata loaders honour this for fast partial reads. | |
| data_sample_n | No | Optional uniform random subsample size (seed=0, deterministic) — useful on huge panels. | |
| drop_controls | No | Baseline controls to drop (one-by-one) for sensitivity. | |
| winsor_levels | No | Winsorization percentiles, e.g. ``[0.01, 0.05]``. | |
| extra_controls | No | Additional controls to add (one-by-one) beyond baseline. |
Output Schema
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||