StatsPAI

sdid

Read-only

Estimate synthetic difference-in-differences treatment effects for one treated unit or block using unit and time weights. Compute placebo, bootstrap, or jackknife standard errors on long panel data.

Instructions

Synthetic difference-in-differences (Arkhangelsky et al. 2021): unit and time weights, one treated unit or block. Long panel; placebo / bootstrap / jackknife SEs. Validation: certified parity evidence.

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": {
      "time": "<time>",
      "unit": "<unit>",
      "y": "<y>"
    },
    "name": "sdid"
  }
}

Input Schema

Name	Required	Description	Default
`y`	Yes	Outcome column.
`seed`	No
`time`	Yes
`unit`	Yes
`treat`	No	0/1 treatment column (block designs).
`detail`	No	Payload depth; agent is the fullest.	agent
`data_id`	No	Data handle (d_...) from load_data; replaces data_path.
`data_csv`	No	Small inline table as CSV text.
`as_handle`	No	Cache the fit and return its result_id.
`data_path`	No	Absolute path of a data file on the server.
`result_id`	No	Handle of an earlier fit made with as_handle=true.
`se_method`	No	'noinference' needs treat=.	placebo
`treat_time`	No	First treated period.
`treat_unit`	No	Treated unit value (or use treat=).
`data_records`	No	Small inline table as row objects.
`max_output_bytes`	No	Result byte budget (default 262144; 0 = none).

Output Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Nov 12, 2026

TDQS

A3.5/5.0

Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered externally. The description adds the method family, the panel shape requirement, and the available inference methods (placebo/bootstrap/jackknife), but says nothing about the handle/caching mechanism or the byte budget that materially shape behavior for a 16-parameter tool.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three compact clauses that are front-loaded with the method identity and scope, and there is no wasted filler. The closing fragment 'Validation: certified parity evidence' is vague jargon that earns little of its space for an executing agent.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a fit tool with an output schema and 81% schema coverage, the description supplies the method identity, applicable data shape, and inference options, which is enough to call it correctly. The main omission is any signal about the data-handle versus inline-CSV versus result_id input paths, though the schema descriptions carry that load.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 81%, so the schema already documents most parameters including the SE enum. The description only echoes the SE options and the single-treated-unit constraint already implied by treat_unit; it adds no meaning for the five data-input parameters or for as_handle/result_id.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific estimator by name with its citation and what it produces ('unit and time weights'), plus scope ('one treated unit or block'). That is clearly more than a restatement of the name, but it never contrasts itself with the sibling estimators (did, callaway_santanna, bacon_decomposition, pipeline_did) that an agent would otherwise have to disambiguate by schema inspection.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

'Long panel' and 'one treated unit or block' imply the data conditions under which this estimator applies, which is real guidance. There is no explicit when-to-use/when-not-to-use statement and no named alternative for the case of multiple treated units or short panels, leaving the routing decision to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.