NMAR0.1.2 package

Estimation under not Missing at Random Nonresponse

apply_nmar_scaling

Apply scaling to a matrix using a recipe

bootstrap_variance.data.frame

Bootstrap for IID data frames

bootstrap_variance.default

Default dispatch

bootstrap_variance

Bootstrap variance estimation module

bootstrap_variance.survey.design

Bootstrap for survey designs

bootstrap_variance.svyrep.design

Replicate-weight designs not supported

coef.nmar_result

Default coefficients for NMAR results

coef.summary_nmar_result

Coefficient table for summary objects

compute_weighted_stats

Compute mean and standard deviation

confint.nmar_result

Wald confidence interval for NMAR results

confint.summary_nmar_result

Confidence intervals for summary objects

constraint_summaries

Constraint summaries for diagnostics

create_nmar_scaling_recipe

Build a scaling recipe from one or more design matrices

create_verboser

Create Verbose Printer Factory

el_assert_no_offset

Assert that terms object lacks offsets

el_augment_strata_aux

Strata augmentation for survey designs

el_build_equation_system_survey

Empirical likelihood equations for survey designs

el_build_equation_system

Empirical likelihood estimating equations for SRS

el_build_jacobian_survey

Empirical likelihood analytical jacobian for survey designs

el_build_jacobian

Empirical likelihood analytical jacobian for srs

el_build_result

Build EL result object

el_build_start

Build starting values

el_check_auxiliary_inconsistency_matrix

Check auxiliary means consistency against respondents sample support.

el_compute_diagnostics

Compute diagnostics

el_compute_variance

Variance driver

el_core_eta_state

Core computations

el_denominator

Compute denominator

el_engine

Empirical likelihood engine for NMAR

el_estimator_core

Core of the empirical likelihood estimator

el_extract_strata_factor

Extract strata factor

el_lambda_W

Compute lambda_W

el_log_banner

Log a step banner line

el_log_data_prep

Log data prep summary

el_log_detailed_diagnostics

Log detailed diagnostics

el_log_final

Log final summary

el_log_solver_config

Log solver configuration

el_log_solver_result

Log solver termination status

el_log_solving

Log a short solver progress note

el_log_start_values

Log starting values

el_log_trace

Log a short trace message with the chosen level

el_log_variance_header

Log variance header and result

el_log_weight_diagnostics

Log weight diagnostics

el_masses

Compute probability masses

el_mean

Compute the mean

el_prepare_inputs

Input preprocessing

el_prepare_nleqslv

Prepare nleqslv args

el_resolve_auxiliaries

Auxiliary design computation

el_run_solver

Solver orchestration

el_validate_design_spec

Validate design dimensions

el_validate_matrix

Validate matrix columns for NA and zero variance

el.data.frame

Empirical likelihood for data frames

el

Empirical likelihood estimator

el.survey.design

Empirical likelihood estimator for survey designs

enforce_nonneg_weights

Enforce nonnegativity of weights

engine_config

Extract engine configuration

engine_name

Canonical engine name

exptilt_engine

Exponential tilting engine for NMAR

exptilt_nonparam_engine

Nonparametric exponential tilting engine for NMAR

exptilt_nonparam

Nonparametric Exponential Tilting (Internal Generic)

exptilt

Exponential tilting estimator

extract_nleqslv_top

Extract top-level nleqslv arguments from a control-like list

fitted.nmar_result

Default fitted values for NMAR results

format.nmar_engine

Formatter for engines

formula.nmar_result

Default formula for NMAR results

generate_conditional_density

Generate conditional density

glance.nmar_result

Glance summary for NMAR results

logit_family

Construct logit response family

merge_nleqslv_top

Prefer explicit solver_args over control-provided top-level args

new_nmar_engine_el

Construct EL Engine Object

new_nmar_result_el

Construct EL Result Object

new_nmar_result

Construct for result objects

nmar_fmt_num

Format a number with fixed decimal places using nmar.digits

nmar_format_call_line

Format an abridged call line for printing

nmar_get_digits

Resolve global digits setting for printing

nmar_get_el_denom_floor

EL denominator floor

nmar_get_numeric_settings

NMAR numeric settings

nmar_result_get_estimate

Internal helpers for nmar_result objects

nmar

Not Missing at Random Estimation

prepare_nmar_scaling

Prepare scaled matrices and moments

print.nmar_engine

Print method for engines

print.nmar_result_el

Print method for EL results

print.nmar_result_exptilt

Print method for Exponential Tilting results (engine-specific)

print.nmar_result

Print method for nmar_result

print.summary_nmar_result

Print method for summary.nmar_result

probit_family

Construct probit response family

run_engine.nmar_engine_el

Run method for EL engine

sanitize_nleqslv_control

Parse nleqslv control list for compatibility

scale_aux_multipliers

Map unscaled auxiliary multipliers to scaled space

scale_coefficients

Map unscaled coefficients to scaled space

se

Extract standard error for NMAR results

shared_weighted_gram

Weighted linear algebra

summary.nmar_result_el

Summary method for EL results

summary.nmar_result_exptilt

Summary method for Exponential Tilting results (engine-specific)

summary.nmar_result

Summary method for nmar_result

tidy.nmar_result

Tidy summary for NMAR results

trim_weights

Trim weights by capping and proportional redistribution

unscale_coefficients

Unscale coefficients and covariance

validate_and_apply_nmar_scaling

Validate and apply scaling for engines

validate_data

Validate Data for NMAR Analysis

validate_nleqslv_top

Validate top-level nleqslv arguments and coerce invalid to defaults

validate_nmar_engine_el

Validate EL Engine Settings

validate_nmar_result

Validate nmar_result

vcov.nmar_result

Variance-covariance for NMAR results

weights.nmar_result

Extract weights from an nmar_result

Methods to estimate finite-population parameters under nonresponse that is not missing at random (NMAR, nonignorable). Incorporates auxiliary information and user-specified response models, and supports independent samples and complex survey designs via objects from the 'survey' package. Provides diagnostics and optional variance estimates. For methodological background see Qin, Leung and Shao (2002) <doi:10.1198/016214502753479338> and Riddles, Kim and Im (2016) <doi:10.1093/jssam/smv047>.

  • Maintainer: Maciej Beresewicz
  • License: MIT + file LICENSE
  • Last published: 2026-02-05