Causal Decision Agent

Overview Schema Source Related Servers Discussions Admin

multi_outcome_synth

Read-only

Estimates causal effects on multiple outcomes using synthetic control, creating a donor-weighted counterfactual to quantify intervention impact with placebo-based inference.

Instructions

Multiple Outcomes Synthetic Control Method (Sun 2023). Assumptions: A convex (or regularized) combination of donor units reproduces the treated unit's pre-treatment outcome path; No interference: the treatment does not affect the donor units (SUTVA); No anticipation before the treatment date. Pre-conditions: Panel of one or more treated units plus an untreated donor pool, observed over time; Pre-treatment window long enough to fit donor weights (rule of thumb: more pre-periods than donors used); Outcome observed for every unit in every period. Failure modes: Large pre-treatment RMSPE -- the synthetic unit fails to track the treated unit before treatment -> Add donors / predictors, lengthen the pre-period, or use a bias-corrected estimator (sdid, augsynth); Placebo / permutation inference shows the estimate is not extreme relative to donors -> Report the placebo distribution honestly; the effect may not be distinguishable from noise. Alternatives: sp.sdid, sp.augsynth, sp.gsynth, sp.callaway_santanna. Typical minimum N: 15.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`time`	Yes	Time period column.
`unit`	Yes	Unit identifier column.
`alpha`	No	Significance level for confidence intervals and joint test.
`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
`method`	No	Weight-estimation strategy. * ``'concatenated'`` -- stack all K standardised outcome panels vertically and solve one quadratic programme. * ``'averaged'`` -- standardise each outcome, average across K, then solve SCM on the mean series.	concatenated
`placebo`	No	Run in-space placebo permutations for inference (each donor is pretended to be treated in turn).
`outcomes`	Yes	Column names for the K outcome variables.
`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.
`standardize`	No	Standardise each outcome to zero mean / unit variance before stacking or averaging (strongly recommended when outcome scales differ).
`data_columns`	No	Optional column projection. Parquet/Feather/Stata loaders honour this for fast partial reads.
`penalization`	No	Ridge-type penalty added to the diagonal of the donor cross-product matrix (``penalization * I``). Helps when donors are collinear.
`treated_unit`	Yes	Value identifying the treated unit.
`data_sample_n`	No	Optional uniform random subsample size (seed=0, deterministic) — useful on huge panels.
`treatment_time`	Yes	First treatment period (inclusive).

Output Schema

TableJSON Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Sep 15, 2026

TDQS

A4.3/5.0

Behavior5/5

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

Beyond the readOnlyHint annotation, the description discloses key causal assumptions (SUTVA, no anticipation), data-shape requirements, failure modes (large pre-treatment RMSPE, placebo inference), and a minimum-N heuristic. This gives an agent the behavioral context needed to anticipate invalid results or misleading estimates.

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

Conciseness5/5

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

The description is dense but well organized: method identification, assumptions, pre-conditions, failure modes, alternatives, and minimum sample size. Each block carries decision-relevant information, and there is no filler or redundant restatement of the tool name.

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

Completeness5/5

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

For a complex causal-estimation tool, the description covers what the method is, when it applies, what can go wrong, and what else to try. An output schema exists, so the absence of return-format details is acceptable. The definition is complete enough for an agent to select and invoke the tool correctly.

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?

The input schema already describes all 16 parameters with 100% coverage, so the baseline is 3. The description adds high-level context about standardization and bias-corrected alternatives but does not materially explain individual parameters beyond what the schema provides. No gap is severe enough to lower the score.

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?

The description identifies the tool as the multiple-outcomes variant of the synthetic control method and cites the method source (Sun 2023), which makes the core purpose clear. It also names several sibling alternatives, helping distinguish it from single-outcome synth or difference-in-differences estimators. It stops short of an explicit verb like 'estimates' or 'constructs', which keeps it just below a top score.

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

Usage Guidelines4/5

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

The description gives concrete pre-conditions: a panel with treated units and donors, sufficient pre-periods, and complete outcome data. It also lists failure modes that signal when the method is inappropriate and names alternatives such as sp.sdid and sp.augsynth. It does not provide explicit decision rules for when to choose each alternative, but the context is strong enough for an agent to make a reasonable selection.

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

Deploy Server