stats.lm

Linear regression by least squares, with R's lm(). Name the model with a formula, or with a target and a column selector for the features. Writes a fitted SpadeModel that models.predict scores without R, the coefficient table, and the fit statistics, with optional cross-validation.

The model records the design columns R produced -- treatment coding, the full coding R uses without an intercept, interactions -- so prediction reproduces them exactly. Formula variables are limited to columns, log(), log1p(), sqrt(), exp() and I() arithmetic; how they combine (:, *, ^, + 0) is not restricted. For tables too large for memory, models.linear fits the same model by streaming.

Kind: standard
Network: no

Inputs🔗

Name	Type	Description
`data`	file (Parquet or CSV)	Training table: one row per observation, with the target, the predictor columns, and any weights or id columns.
`formula`	string	Model formula, e.g. `ESTIMATE ~ B5 + log(B25) + FORTYPE + 0`. Use instead of `target` and `features`. Default empty.
`target`	string	The response column (when no formula is given). Default empty.
`features`	string	Column selector for the predictors (when no formula is given): names, globs and exclusions, e.g. `B,MB,-B0`. Default empty.
`intercept`	boolean	Include an intercept when building the model from `target` and `features`. With a formula, write `+ 0` in the formula instead. Default `true`.
`id_columns`	string	Column selector for identifier columns, never used as predictors; kept as text when reading CSV. Default empty.
`weights`	string	Column of non-negative observation weights. Empty fits unweighted.
`categorical`	string	Column selector for predictors to treat as categorical even though they are stored as numbers. Default empty.
`na_action`	string	`fail` (default, naming the columns) or `drop`: what to do with a row that has a missing value in a model column.
`seed`	number	Seed for cross-validation fold assignment. Default `0`.
`cv_folds`	number	K-fold cross-validation (`0` for none). The model is always fitted on all rows. Default `0`.
`cv_group`	string	Column whose values are kept together in one fold (e.g. a county id), so spatially related rows do not leak between folds. Default empty.

Outputs🔗

Name	Type	Description
`model`	file (SpadeModel)	The fitted model (family `linear`), for `models.predict`.
`coefficients`	file (CSV)	One row per design column: `term`, `estimate`, `std_error`, `statistic`, `p_value`.
`fit`	json	Rows used and dropped, the arguments that shaped the fit, and `metrics`: `rmse`, `mae`, `r_squared`, `adj_r_squared`, `aic`, `bic`, `sigma`, `df_residual`, and `cv_rmse`, `cv_mae`, `cv_r_squared` when cross-validated.

Example🔗

- id: "@fit"
  name: stats.lm
  inputs:
    - "@plots"
  args:
    formula: "biomass ~ canopy_cover + log(stand_age) + forest_type"
    cv_folds: 5
    cv_group: "county"