GLM

Defined in: ds/src/stats/estimators/GLM.js:19

Extends

• Estimator

Constructors

Constructor

new GLM(params?): GLM

Defined in: ds/src/stats/estimators/GLM.js:38

Parameters

params?

Model parameters

alpha

number

Significance level for confidence intervals (default: 0.05 for 95% CIs)

compress

boolean

Compress model to save memory (default: false)

dispersion

string | number

Dispersion estimation: ‘estimate’, ‘fixed’, or numeric

family

string

GLM family (gaussian, binomial, poisson, gamma, inverse_gaussian, negative_binomial)

intercept

boolean

Include intercept term (default: true)

keepFittedValues

boolean

Keep fitted values and residuals (default: true)

link

string

Link function (default: canonical link for family)

maxIter

number

Maximum iterations (default: 100)

multiclass

string

Multiclass strategy: ‘ovr’ (one-vs-rest), ‘multinomial’ (softmax), or null (binary/regression)

randomEffects

any

Random effects specification {intercept: […], slopes: {…}}

regularization

any

Regularization {alpha, l1_ratio}

theta

number

Theta parameter for negative binomial (default: 1)

tol

number

Convergence tolerance (default: 1e-8 for GLM, 1e-6 for GLMM)

warnLargeDataset

boolean

Warn about large datasets in browser (default: true)

warnOnNoConvergence

boolean

Warn if model doesn’t converge (default: true)

Returns

GLM

Overrides

Estimator.constructor

Properties

_classes

_classes: any[]

Defined in: ds/src/stats/estimators/GLM.js:73

---

_columnsX

_columnsX: any

Defined in: ds/src/stats/estimators/GLM.js:78

---

_columnY

_columnY: any

Defined in: ds/src/stats/estimators/GLM.js:79

---

_formula

_formula: any

Defined in: ds/src/stats/estimators/GLM.js:208

---

_indicatorColumns

_indicatorColumns: string[]

Defined in: ds/src/stats/estimators/GLM.js:382

---

_isMixed

_isMixed: boolean

Defined in: ds/src/stats/estimators/GLM.js:77

---

_isMulticlass

_isMulticlass: boolean

Defined in: ds/src/stats/estimators/GLM.js:75

---

_isMultinomial

_isMultinomial: boolean

Defined in: ds/src/stats/estimators/GLM.js:448

---

_isMultiOutput

_isMultiOutput: boolean

Defined in: ds/src/stats/estimators/GLM.js:76

---

_model

_model: any

Defined in: ds/src/stats/estimators/GLM.js:71

---

_models

_models: object

Defined in: ds/src/stats/estimators/GLM.js:72

---

_state

_state: object

Defined in: ds/src/core/estimators/estimator.js:27

Inherited from

Estimator._state

---

_targetNames

_targetNames: any

Defined in: ds/src/stats/estimators/GLM.js:74

---

_warnings

_warnings: any[]

Defined in: ds/src/core/estimators/estimator.js:29

Inherited from

OneSampleTTest._warnings

---

fitted

fitted: boolean

Defined in: ds/src/core/estimators/estimator.js:25

Inherited from

OneSampleTTest.fitted

---

params

params: object

Defined in: ds/src/stats/estimators/GLM.js:42

alpha

alpha: number

compress

compress: boolean

dispersion

dispersion: string | number

family

family: string

intercept

intercept: boolean

keepFittedValues

keepFittedValues: boolean

link

link: string

maxIter

maxIter: number

multiclass

multiclass: string

randomEffects

randomEffects: any

regularization

regularization: any

theta

theta: number

tol

tol: number

warnLargeDataset

warnLargeDataset: boolean

warnOnNoConvergence

warnOnNoConvergence: boolean

Inherited from

OneSampleTTest.params

Accessors

coefficients

Get Signature

get coefficients(): any

Defined in: ds/src/stats/estimators/GLM.js:85

Get coefficients (for backward compatibility with lm interface)

Returns

any

---

intercept

Get Signature

get intercept(): boolean

Defined in: ds/src/stats/estimators/GLM.js:93

Get intercept flag (for backward compatibility)

Returns

boolean

Methods

_extractColumn()

_extractColumn(columnName, _data, rows): any

Defined in: ds/src/stats/estimators/GLM.js:713

Extract a column from table data

Parameters

columnName

any

_data

any

rows

any

Returns

any

---

_getCoefLabels()

_getCoefLabels(): any[]

Defined in: ds/src/stats/estimators/GLM.js:1632

Get coefficient labels

Returns

any[]

---

_parseRandomEffects()

_parseRandomEffects(opts, rows): object

Defined in: ds/src/stats/estimators/GLM.js:686

Parse random effects specification from table-style input

Parameters

opts

any

rows

any

Returns

object

intercept

intercept: any

slopes

slopes: object

---

_prepareArgsForFit()

_prepareArgsForFit(args?): { columns?: undefined; columnsX: any[]; prepared: boolean; raw?: undefined; rows: any[]; X: any[][]; y: any[]; } | { columns: any[]; columnsX?: undefined; prepared: boolean; raw?: undefined; rows: any[]; X: any[][]; y?: undefined; } | { columns?: undefined; columnsX?: undefined; prepared?: undefined; raw: any[]; rows?: undefined; X?: undefined; y?: undefined; }

Defined in: ds/src/core/estimators/estimator.js:367

Convenience helper: parse arguments passed to fit/predict/transform.

Supports declarative table-style inputs:

• fit({ X, y, data, omit_missing })
• fit({ data, columns, … })

Returns an object { X, y, prepared, rows } where X/y are numeric arrays if preparation was required, otherwise returns the original values.

Note: this helper only prepares numeric matrices/vectors using core table utilities; it does not perform encoding of categorical predictors.

Parameters

args?

any[] = []

Returns

{ columns?: undefined; columnsX: any[]; prepared: boolean; raw?: undefined; rows: any[]; X: any[][]; y: any[]; } | { columns: any[]; columnsX?: undefined; prepared: boolean; raw?: undefined; rows: any[]; X: any[][]; y?: undefined; } | { columns?: undefined; columnsX?: undefined; prepared?: undefined; raw: any[]; rows?: undefined; X?: undefined; y?: undefined; }

Inherited from

Estimator._prepareArgsForFit

---

_repr_html_()

_repr_html_(): string

Defined in: ds/src/stats/estimators/GLM.js:1138

Jupyter notebook display support Returns HTML representation for better notebook rendering

Returns

string

Overrides

Estimator._repr_html_

---

_summaryGLM()

_summaryGLM(alpha?): string

Defined in: ds/src/stats/estimators/GLM.js:1161

Format GLM summary

Parameters

alpha?

number = 0.05

Returns

string

---

_summaryGLMHTML()

_summaryGLMHTML(): string

Defined in: ds/src/stats/estimators/GLM.js:1373

Format GLM summary as HTML for Jupyter

Returns

string

---

_summaryGLMM()

_summaryGLMM(alpha?): string

Defined in: ds/src/stats/estimators/GLM.js:1448

Format GLMM summary (lme4-style, no p-values)

Parameters

alpha?

number = 0.05

Returns

string

---

_summaryGLMMHTML()

_summaryGLMMHTML(): string

Defined in: ds/src/stats/estimators/GLM.js:1528

Format GLMM summary as HTML for Jupyter

Returns

string

---

_summaryMulticlass()

_summaryMulticlass(alpha?): string

Defined in: ds/src/stats/estimators/GLM.js:1225

Format multiclass GLM summary

Parameters

alpha?

number = 0.05

Returns

string

---

_summaryMultiOutput()

_summaryMultiOutput(alpha?): string

Defined in: ds/src/stats/estimators/GLM.js:1248

Format multi-output GLM summary

Parameters

alpha?

number = 0.05

Returns

string

---

_summaryTrueMultinomial()

_summaryTrueMultinomial(alpha?): string

Defined in: ds/src/stats/estimators/GLM.js:1289

Format true multinomial logistic regression summary

Parameters

alpha?

number = 0.05

Returns

string

---

clearWarnings()

clearWarnings(): void

Defined in: ds/src/core/estimators/estimator.js:139

Clear all warnings

Returns

void

Inherited from

Estimator.clearWarnings

---

confint()

confint(alpha?): any[]

Defined in: ds/src/stats/estimators/GLM.js:1036

Compute confidence intervals for coefficients

Parameters

alpha?

number = null

Significance level (default: 0.05 for 95% CIs)

Returns

any[]

Array of {lower, upper} for each coefficient

---

fit()

fit(…args): GLM

Defined in: ds/src/stats/estimators/GLM.js:108

Fit the GLM or GLMM

Supports multiple calling conventions:

• fit(X, y)
• fit(X, y, weights, offset)
• fit({ X, y, data })
• fit({ X, y, groups, data }) for mixed models
• fit(‘y ~ x1 + x2’, data) - R-style formula
• fit({ formula: ‘y ~ x1 + x2’, data }) - formula in object

Parameters

args

…any[]

Returns

GLM

Overrides

Estimator.fit

---

getMemoryUsage()

getMemoryUsage(): string

Defined in: ds/src/core/estimators/estimator.js:97

Get memory usage in human-readable format

Returns

string

Memory usage string (e.g., “2.3 MB” or “145 KB”)

Inherited from

Estimator.getMemoryUsage

---

getParams()

getParams(): any

Defined in: ds/src/core/estimators/estimator.js:294

Get a shallow copy of parameters.

Returns

any

Inherited from

Estimator.getParams

---

getState()

getState(): any

Defined in: ds/src/core/estimators/estimator.js:65

Get comprehensive model state

Returns

any

State information including fitted status, memory estimate, warnings

Inherited from

Estimator.getState

---

getWarnings()

getWarnings(): any[]

Defined in: ds/src/core/estimators/estimator.js:124

Get all warnings

Returns

any[]

Array of warning objects

Inherited from

Estimator.getWarnings

---

getWarningsByType()

getWarningsByType(type): any[]

Defined in: ds/src/core/estimators/estimator.js:148

Get warnings of a specific type

Parameters

type

string

Warning type

Returns

any[]

Filtered warnings

Inherited from

Estimator.getWarningsByType

---

hasWarnings()

hasWarnings(): boolean

Defined in: ds/src/core/estimators/estimator.js:132

Check if model has warnings

Returns

boolean

Inherited from

Estimator.hasWarnings

---

isFitted()

isFitted(): boolean

Defined in: ds/src/core/estimators/estimator.js:36

Check if model is fitted

Returns

boolean

Inherited from

Estimator.isFitted

---

predict()

predict(X, options?): any[]

Defined in: ds/src/stats/estimators/GLM.js:941

Predict from the fitted model

Parameters

X

any

Predictors or table-style object

options?

Prediction options

allowNewGroups

boolean

For GLMM: allow new groups (default: true)

interval

boolean

Compute confidence intervals (default: false)

level

number

Confidence level (default: 0.95)

type

string

Prediction type: ‘link’, ‘response’, ‘class’ (multiclass), ‘proba’ (multiclass)

Returns

any[]

Predictions

Overrides

Estimator.predict

---

pvalues()

pvalues(): number[]

Defined in: ds/src/stats/estimators/GLM.js:1059

Compute p-values for coefficients (Wald test)

Note: For mixed models, p-values are controversial and should be interpreted with caution. Prefer confidence intervals.

Returns

number[]

P-values for each coefficient

---

save()

save(): string

Defined in: ds/src/core/estimators/estimator.js:329

Save model to JSON string

Returns

string

JSON representation of the model

Inherited from

Estimator.save

---

score()

score(yTrue, yPred, …args): number

Defined in: ds/src/stats/estimators/GLM.js:1660

Score the model (R² for regression families, accuracy for binomial/multiclass)

Parameters

yTrue

any

yPred

any

args

…any[]

Returns

number

---

setParams()

setParams(params?): GLM

Defined in: ds/src/core/estimators/estimator.js:285

Set parameters (mutates instance).

Parameters

params?

any = {}

Returns

GLM

Inherited from

Estimator.setParams

---

summary()

summary(options?): string

Defined in: ds/src/stats/estimators/GLM.js:1015

Get model summary (lme4-style for mixed models)

Parameters

options?

Summary options

alpha

number

Significance level for CIs (default: from constructor)

Returns

string

---

toJSON()

toJSON(): object

Defined in: ds/src/stats/estimators/GLM.js:1709

Serialize to JSON

Returns

object

__class__

__class__: string = 'GLM'

classes

classes: any[]

columnsX

columnsX: any

columnY

columnY: any

fitted

fitted: boolean

isMixed

isMixed: boolean

isMulticlass

isMulticlass: boolean

isMultinomial

isMultinomial: boolean

isMultiOutput

isMultiOutput: boolean

model

model: any

models

models: object

Index Signature

[k: string]: any

params

params: object

params.alpha

alpha: number

params.compress

compress: boolean

params.dispersion

dispersion: string | number

params.family

family: string

params.intercept

intercept: boolean

params.keepFittedValues

keepFittedValues: boolean

params.link

link: string

params.maxIter

maxIter: number

params.multiclass

multiclass: string

params.randomEffects

randomEffects: any

params.regularization

regularization: any

params.theta

theta: number

params.tol

tol: number

params.warnLargeDataset

warnLargeDataset: boolean

params.warnOnNoConvergence

warnOnNoConvergence: boolean

targetNames

targetNames: any

Overrides

Estimator.toJSON

---

transform()

transform(): void

Defined in: ds/src/core/estimators/estimator.js:431

Transform should be implemented by transformers.

Returns

void

Inherited from

Estimator.transform

---

fromJSON()

static fromJSON(obj): GLM

Defined in: ds/src/stats/estimators/GLM.js:1734

Deserialize from JSON

Parameters

obj

any

Returns

GLM

Overrides

Estimator.fromJSON

---

load()

static load(jsonString): Estimator

Defined in: ds/src/core/estimators/estimator.js:346

Load model from JSON string

Parameters

jsonString

string

JSON representation

Returns

Estimator

Reconstructed estimator instance

Inherited from

Estimator.load