{ "tools": [ { "name": "explore_descriptive_table", "description": "Report descriptive statistics for the numeric/logical variables of ``df``.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: descriptive_stats.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "stats": { "type": "array", "items": { "type": "string" }, "description": "Statistics to display, in order, chosen from ``N``, ``Mean``, ``Std. dev.``, ``Min.``, ``25 %``, ``Median``, ``75 %``, ``Max.``. Defaults to ``Mean``, ``Std. dev.``, ``Median``, ``Min.``, ``Max.``. (The returned ``.df`` always carries all eight statistics regardless of this selection.)", "default": [ "Mean", "Std. dev.", "Median", "Min.", "Max." ] }, "digits": { "oneOf": [ { "type": "integer" }, { "type": "object" } ], "description": "Number of decimals for the displayed statistics: a single ``int`` applied to all, or a ``{statistic: decimals}`` mapping for per-statistic overrides (``N`` is always shown as an integer).", "default": 3 }, "periods": { "type": "array", "items": { "type": "string" }, "description": "Periods to show as sub-columns in the by-period layout. ``None`` (default) shows the first and last period; otherwise the listed period values (those not present are dropped with a warning). Ignored when no time dimension is summarized." }, "entity": { "type": "string", "description": "Optional panel identifiers (defaulting to those declared via :func:`expdpy.set_panel`). A resolved ``time`` drives the by-period layout; when both resolve, a note also reports the panel dimensions. For the within/between split of each variable see :func:`expdpy.explore_xtsum_table`. If omitted, the panel declared by set_panel() on the data is used." }, "time": { "type": "string", "description": "If omitted, the panel declared by set_panel() on the data is used." } }, "required": [ "data" ], "additionalProperties": false } }, { "name": "explore_correlation_table", "description": "Correlation table with Pearson above and Spearman below the diagonal.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: pearson, correlation_vs_causation.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "digits": { "type": "integer", "description": "Number of decimals to display (``0 < digits < 5``).", "default": 2 }, "bold": { "type": "number", "description": "Correlations with a p-value below ``bold`` are shown in bold. Set to ``0`` to disable. Must satisfy ``0 <= bold < 1``.", "default": 0.05 } }, "required": [ "data" ], "additionalProperties": false } }, { "name": "explore_scatter_plot", "description": "Scatter plot of ``y`` against ``x`` with optional aesthetics and a LOESS smoother.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: correlation_vs_causation.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "x": { "type": "string", "description": "Column names for the axes (both must be numeric). When omitted, they default to the declared roles (:func:`expdpy.set_roles`): ``x`` to the first covariate and ``y`` to the main outcome." }, "y": { "type": "string" }, "color": { "type": "string", "description": "Optional column mapped to marker color (numeric -> colorbar, otherwise discrete)." }, "size": { "type": "string", "description": "Optional numeric column mapped to marker size." }, "loess": { "type": "integer", "enum": [ 0, 1, 2 ], "description": "``0`` no smoother, ``1`` unweighted LOESS, ``2`` LOESS weighted by ``size``.", "default": 0 }, "alpha": { "type": "number", "description": "Marker opacity. If ``None``, a sample-size-based default is used." }, "entity": { "type": "string", "description": "Cross-sectional (unit) identifier. Defaults to the panel ``entity`` declared via :func:`expdpy.set_panel`. Only used when ``connect=True``. If omitted, the panel declared by set_panel() on the data is used." }, "time": { "type": "string", "description": "Time identifier used to order each unit's trajectory when ``connect=True``. Defaults to the panel ``time``. If omitted, the panel declared by set_panel() on the data is used." }, "connect": { "type": "boolean", "description": "If ``True`` (and an ``entity`` is available), draw a faint line connecting each unit's points in time order \u2014 turning the scatter into a panel trajectory plot.", "default": false } }, "required": [ "data" ], "additionalProperties": false } }, { "name": "explore_trend_plot", "description": "Line-plot the mean (with standard-error bars) of variables over time.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: time_trends.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "var": { "type": "array", "items": { "type": "string" }, "description": "Variables to plot. Defaults to all numeric columns other than ``time``/``entity``." }, "time": { "type": "string", "description": "Column name of the time identifier. Defaults to the panel ``time`` declared via :func:`expdpy.set_panel`. If omitted, the panel declared by set_panel() on the data is used." }, "entity": { "type": "string", "description": "Column name of the cross-sectional (unit) identifier. Only used when ``spaghetti=True``. Defaults to the panel ``entity`` declared via :func:`expdpy.set_panel`. If omitted, the panel declared by set_panel() on the data is used." }, "spaghetti": { "type": "boolean", "description": "If ``True`` (and a single ``var`` and an ``entity`` are available), draw a faint per-unit trajectory backdrop behind the mean line. For richer per-unit views see :func:`expdpy.explore_spaghetti_plot`.", "default": false } }, "required": [ "data" ], "additionalProperties": false } }, { "name": "explore_missing_values_plot", "description": "Heatmap of missing-value frequency by variable and panel dimension.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: panel_structure.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "time": { "type": "string", "description": "Time identifier column. Defaults to the panel ``time`` declared via :func:`expdpy.set_panel`. Required when ``by=\"time\"``; must not contain missing values. If omitted, the panel declared by set_panel() on the data is used." }, "entity": { "type": "string", "description": "Cross-sectional (unit) identifier column. Defaults to the panel ``entity``. Required when ``by=\"entity\"``; must not contain missing values. If omitted, the panel declared by set_panel() on the data is used." }, "by": { "type": "string", "enum": [ "time", "entity" ], "description": "Whether to aggregate missingness over ``\"time\"`` periods (the default) or over ``\"entity\"`` units.", "default": "time" }, "no_factors": { "type": "boolean", "description": "If ``True``, limit the plot to numeric/logical variables.", "default": false }, "binary": { "type": "boolean", "description": "If ``True``, show only whether values are missing (any) rather than the fraction.", "default": false } }, "required": [ "data" ], "additionalProperties": false } }, { "name": "explore_panel_structure", "description": "Summarise the panel's balance and coverage, with a unit-by-period presence grid.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: panel_structure.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "entity": { "type": "string", "description": "Cross-sectional (unit) identifier. Defaults to the panel ``entity``. If omitted, the panel declared by set_panel() on the data is used." }, "time": { "type": "string", "description": "Time identifier. Defaults to the panel ``time``. If omitted, the panel declared by set_panel() on the data is used." }, "var": { "type": "string", "description": "Optional variable: when given, a cell counts as \"present\" only if ``var`` is non-missing there (rather than merely a row existing)." }, "max_units": { "type": "integer", "description": "Cap on the number of units drawn in the presence grid (evenly sampled above it).", "default": 200 } }, "required": [ "data" ], "additionalProperties": false } }, { "name": "analyze_regression_table", "description": "Build a regression table of one or more OLS models.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: ols, fixed_effects, clustered_se.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "dvs": { "oneOf": [ { "type": "array", "items": { "type": "string" } }, { "type": "string" } ], "description": "Dependent variable name, or a list of names (one per model). Defaults to the declared main outcome (:func:`expdpy.set_roles`) when omitted." }, "idvs": { "oneOf": [ { "type": "array", "items": { "type": "string" } }, { "type": "array", "items": { "type": "array", "items": { "type": "string" } } } ], "description": "Independent variable names. For multiple models, a list of lists. Defaults to the declared covariates (:func:`expdpy.set_roles`) when omitted." }, "feffects": { "oneOf": [ { "type": "array", "items": { "type": "string" } }, { "type": "array", "items": { "type": "array", "items": { "type": "string" } } } ], "description": "Fixed-effects variable names (per model when multiple models are given)." }, "clusters": { "oneOf": [ { "type": "array", "items": { "type": "string" } }, { "type": "array", "items": { "type": "array", "items": { "type": "string" } } } ], "description": "Cluster variable names for clustered standard errors (per model when multiple)." }, "byvar": { "type": "string", "description": "A categorical variable to estimate the single model separately by. Only valid with a single dependent variable. Levels with too few observations to estimate the model (``n <= len(idvs) + 1``) are skipped." } }, "required": [ "data" ], "additionalProperties": false } }, { "name": "analyze_panel_table", "description": "Estimate pooled / between / fixed / random-effects models side by side.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: fixed_effects, random_effects, hausman.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "dv": { "type": "string", "description": "Dependent variable name." }, "idvs": { "oneOf": [ { "type": "array", "items": { "type": "string" } }, { "type": "string" } ], "description": "Independent variable name(s)." }, "entity": { "type": "string", "description": "The cross-section and time identifiers. Default to the panel declared via :func:`expdpy.set_panel` / :func:`expdpy.set_labels` (``set_panel=True``). If omitted, the panel declared by set_panel() on the data is used." }, "time": { "type": "string", "description": "If omitted, the panel declared by set_panel() on the data is used." }, "models": { "type": "array", "items": { "type": "string", "enum": [ "pooled", "between", "fe", "re" ] }, "description": "Which estimators to include, in order.", "default": [ "pooled", "between", "fe", "re" ] }, "cov_type": { "type": "string", "enum": [ "clustered", "robust", "unadjusted" ], "description": "Covariance estimator: ``\"clustered\"`` (default), ``\"robust\"`` or ``\"unadjusted\"``.", "default": "clustered" }, "cluster_entity": { "type": "boolean", "description": "Cluster by entity when ``cov_type=\"clustered\"``.", "default": true } }, "required": [ "data", "dv", "idvs" ], "additionalProperties": false } }, { "name": "analyze_iv_regression", "description": "Fit an instrumental-variables (2SLS) regression with a weak-instrument diagnostic.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: instrumental_variables, correlation_vs_causation.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "dv": { "type": "string", "description": "Dependent (outcome) variable name." }, "endog": { "oneOf": [ { "type": "array", "items": { "type": "string" } }, { "type": "string" } ], "description": "Endogenous regressor name(s) to be instrumented." }, "instruments": { "oneOf": [ { "type": "array", "items": { "type": "string" } }, { "type": "string" } ], "description": "Excluded instrument name(s). At least as many instruments as endogenous regressors are required (the order condition); more instruments over-identify the model." }, "exog": { "type": "array", "items": { "type": "string" }, "description": "Included exogenous regressor (control) names, if any." }, "feffects": { "type": "array", "items": { "type": "string" }, "description": "Fixed-effects variable names absorbed by pyfixest." }, "clusters": { "type": "array", "items": { "type": "string" }, "description": "Cluster variable name(s) for cluster-robust standard errors." } }, "required": [ "data", "dv", "endog", "instruments" ], "additionalProperties": false } }, { "name": "analyze_event_study", "description": "Estimate and plot an event study for staggered treatment adoption.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: event_study, parallel_trends.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "outcome": { "type": "string", "description": "Outcome variable name." }, "unit": { "type": "string", "description": "Unit (cross-section) identifier. Defaults to the declared panel entity." }, "time": { "type": "string", "description": "Time identifier. Defaults to the declared panel time. If omitted, the panel declared by set_panel() on the data is used." }, "cohort": { "type": "string", "description": "First-treated period for each unit; ``never_treated_value`` marks never-treated units." }, "estimator": { "type": "string", "enum": [ "did2s", "twfe", "saturated", "lpdid" ], "description": "``\"did2s\"`` (Gardner two-stage, the default and robust to heterogeneity), ``\"twfe\"`` (classic two-way fixed effects \u2014 shown for comparison, biased under heterogeneous effects), ``\"saturated\"`` (Sun-Abraham, one curve per cohort) or ``\"lpdid\"`` (local-projections DiD).", "default": "did2s" }, "cluster": { "type": "string", "description": "Cluster variable for standard errors (defaults to ``unit``)." }, "pre_window": { "type": "integer", "description": "Event-time window for ``\"lpdid\"`` (ignored by the other estimators)." }, "post_window": { "type": "integer" }, "never_treated_value": { "type": "integer", "description": "The value of ``cohort`` that marks never-treated units (default ``0``).", "default": 0 } }, "required": [ "data", "outcome", "cohort" ], "additionalProperties": false } }, { "name": "analyze_beta_convergence", "description": "Unconditional and conditional \u03b2-convergence for a panel variable.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: beta_convergence.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "var": { "type": "string", "description": "Numeric variable to analyse (e.g. ``\"log_gdp_pc\"``)." }, "controls": { "oneOf": [ { "type": "array", "items": { "type": "string" } }, { "type": "string" } ], "description": "Optional control name(s). Each enters the conditional model as its **initial-year** value. ``None`` runs the unconditional analysis only." }, "entity": { "type": "string", "description": "Panel identifiers. Default to those declared via :func:`expdpy.set_panel`. If omitted, the panel declared by set_panel() on the data is used." }, "time": { "type": "string", "description": "If omitted, the panel declared by set_panel() on the data is used." }, "start": { "type": "number", "description": "First and last year used to form the growth rate. Default to the earliest and latest year in the panel; only units observed at **both** endpoints are kept." }, "end": { "type": "number" }, "rolling": { "type": "boolean", "description": "When ``True`` (default), also estimate the convergence slope on every fixed-width window of ``window`` periods and return the time path.", "default": true }, "window": { "type": "integer", "description": "Width of the rolling window in **periods** (sorted distinct years). Default ``max(2, n_periods // 2)``." }, "min_obs": { "type": "integer", "description": "Minimum number of units required in the cross-section and in each rolling window.", "default": 10 }, "vcov": { "type": "string", "enum": [ "hetero", "iid" ], "description": "Standard-error type for the reported coefficients: ``\"hetero\"`` (HC1, the default) or ``\"iid\"``. Does not affect the point estimate or the plotted band.", "default": "hetero" } }, "required": [ "data", "var" ], "additionalProperties": false } }, { "name": "analyze_sigma_convergence", "description": "\u03c3-convergence: track and test the cross-sectional dispersion of a panel variable.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: sigma_convergence.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "var": { "type": "string", "description": "Numeric variable whose cross-sectional dispersion is tracked (e.g. ``\"lifeExp\"``). Used as supplied. The Gini index additionally requires non-negative values." }, "entity": { "type": "string", "description": "Panel identifiers. Default to those declared via :func:`expdpy.set_panel`. If omitted, the panel declared by set_panel() on the data is used." }, "time": { "type": "string", "description": "If omitted, the panel declared by set_panel() on the data is used." }, "start": { "type": "number", "description": "Optional first and last period to include. Default to the full range; the retained window must still be balanced." }, "end": { "type": "number" }, "min_periods": { "type": "integer", "description": "Minimum number of periods required to estimate a dispersion trend (at least 3).", "default": 3 }, "vcov": { "type": "string", "enum": [ "hetero", "iid" ], "description": "Standard-error type for the trend coefficients: ``\"hetero\"`` (HC1, the default) or ``\"iid\"``. Does not change the point estimate.", "default": "hetero" } }, "required": [ "data", "var" ], "additionalProperties": false } }, { "name": "analyze_kuznets_waves", "description": "Estimate the extended Kuznets curve (\"Kuznets waves\") across three panel estimators.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.\nSee also concepts: kuznets_waves.", "input_schema": { "type": "object", "properties": { "data": { "description": "The dataset to analyse. Provide ONE of: a bundled dataset name; an absolute path to a .csv/.parquet readable by the server; or inline row records. Panel data uses entity (unit) + time columns.", "oneOf": [ { "type": "object", "title": "BundledDataset", "properties": { "dataset": { "type": "string", "enum": [ "kuznets", "gapminder", "staggered_did", "firms", "productivity", "bolivia112_gdppc", "colonial_origins", "regional_conflict" ] }, "with_labels": { "type": "boolean", "default": true } }, "required": [ "dataset" ], "additionalProperties": false }, { "type": "object", "title": "DataPath", "properties": { "path": { "type": "string", "description": "Absolute path to a .csv or .parquet file." } }, "required": [ "path" ], "additionalProperties": false }, { "type": "object", "title": "InlineRecords", "properties": { "records": { "type": "array", "items": { "type": "object" }, "description": "Inline row records (list of column->value objects); capped by the server to keep token budgets small." } }, "required": [ "records" ], "additionalProperties": false } ] }, "inequality": { "type": "string", "description": "Numeric outcome (an inequality measure such as a Gini). Default ``\"gini_regional\"``.", "default": "gini_regional" }, "development": { "type": "string", "description": "Numeric development regressor, used as supplied (typically *log* GDP per capita); its powers ``g^2 .. g^degree`` are formed internally. Default ``\"log_gdp_pc\"``.", "default": "log_gdp_pc" }, "controls": { "oneOf": [ { "type": "array", "items": { "type": "string" } }, { "type": "string" } ], "description": "Optional covariate name(s) partialled out of the **between** and **within** figures via the Frisch-Waugh-Lovell theorem. They do **not** enter the comparison tables, which are pure polynomial buildups. ``None`` (default) partials out only the fixed effects (in the within figure)." }, "entity": { "type": "string", "description": "Panel identifiers. Default to those declared via :func:`expdpy.set_panel`. If omitted, the panel declared by set_panel() on the data is used." }, "time": { "type": "string", "description": "If omitted, the panel declared by set_panel() on the data is used." }, "degree": { "type": "integer", "description": "Polynomial order in ``[2, 6]`` (default 4, the quartic \"waves\" specification). The tables show ``degree`` nested columns; ``degree=2`` recovers the classic inverted-U test.", "default": 4 }, "vcov": { "type": "string", "enum": [ "hetero", "iid" ], "description": "Standard-error type for the **pooled** and **between** tables: ``\"hetero\"`` (HC1, the default) or ``\"iid\"``. The **within** table always uses standard errors clustered by ``entity``. Does not change any point estimate.", "default": "hetero" }, "n_sample": { "type": "integer", "description": "Number of points drawn in the raw and within scatters (default 1000); the fitted curves always use every row. ``None`` plots every point. The between scatter (one point per entity) is never subsampled.", "default": 1000 }, "seed": { "type": "integer", "description": "Seed for the point-subsampling RNG (default 0), for reproducible figures.", "default": 0 } }, "required": [ "data" ], "additionalProperties": false } }, { "name": "explain", "description": "Return the :class:`Explainer` for a method or concept.\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.", "input_schema": { "type": "object", "properties": { "topic": { "type": "string", "description": "A topic key or alias (see :func:`list_topics`)." }, "lang": { "type": "string", "description": "Language code. Only ``\"en\"`` ships today; the parameter is reserved so that adding translations later is non-breaking.", "default": "en" } }, "required": [ "topic" ], "additionalProperties": false } }, { "name": "list_topics", "description": "Return the sorted list of canonical topic keys (for app menus and docs).\n\nexpdpy reports statistical associations, not proof of causation. Describe results with associational language (\"is associated with\", \"co-moves with\"), not causal language. Panel data uses entity (unit) + time vocabulary; declare it once with set_panel(entity=, time=). Every result exposes interpret() for a plain-language reading and explain(topic) for method caveats; prefer them over ad-hoc narration.", "input_schema": { "type": "object", "properties": {}, "required": [], "additionalProperties": false } } ] }