Estimation under not Missing at Random Nonresponse
Apply scaling to a matrix using a recipe
Bootstrap for IID data frames
Default dispatch
Bootstrap variance estimation module
Bootstrap for survey designs
Replicate-weight designs not supported
Default coefficients for NMAR results
Coefficient table for summary objects
Compute mean and standard deviation
Wald confidence interval for NMAR results
Confidence intervals for summary objects
Constraint summaries for diagnostics
Build a scaling recipe from one or more design matrices
Create Verbose Printer Factory
Assert that terms object lacks offsets
Strata augmentation for survey designs
Empirical likelihood equations for survey designs
Empirical likelihood estimating equations for SRS
Empirical likelihood analytical jacobian for survey designs
Empirical likelihood analytical jacobian for srs
Build EL result object
Build starting values
Check auxiliary means consistency against respondents sample support.
Compute diagnostics
Variance driver
Core computations
Compute denominator
Empirical likelihood engine for NMAR
Core of the empirical likelihood estimator
Extract strata factor
Compute lambda_W
Log a step banner line
Log data prep summary
Log detailed diagnostics
Log final summary
Log solver configuration
Log solver termination status
Log a short solver progress note
Log starting values
Log a short trace message with the chosen level
Log variance header and result
Log weight diagnostics
Compute probability masses
Compute the mean
Input preprocessing
Prepare nleqslv args
Auxiliary design computation
Solver orchestration
Validate design dimensions
Validate matrix columns for NA and zero variance
Empirical likelihood for data frames
Empirical likelihood estimator
Empirical likelihood estimator for survey designs
Enforce nonnegativity of weights
Extract engine configuration
Canonical engine name
Exponential tilting engine for NMAR
Nonparametric exponential tilting engine for NMAR
Nonparametric Exponential Tilting (Internal Generic)
Exponential tilting estimator
Extract top-level nleqslv arguments from a control-like list
Default fitted values for NMAR results
Formatter for engines
Default formula for NMAR results
Generate conditional density
Glance summary for NMAR results
Construct logit response family
Prefer explicit solver_args over control-provided top-level args
Construct EL Engine Object
Construct EL Result Object
Construct for result objects
Format a number with fixed decimal places using nmar.digits
Format an abridged call line for printing
Resolve global digits setting for printing
EL denominator floor
NMAR numeric settings
Internal helpers for nmar_result objects
Not Missing at Random Estimation
Prepare scaled matrices and moments
Print method for engines
Print method for EL results
Print method for Exponential Tilting results (engine-specific)
Print method for nmar_result
Print method for summary.nmar_result
Construct probit response family
Run method for EL engine
Parse nleqslv control list for compatibility
Map unscaled auxiliary multipliers to scaled space
Map unscaled coefficients to scaled space
Extract standard error for NMAR results
Weighted linear algebra
Summary method for EL results
Summary method for Exponential Tilting results (engine-specific)
Summary method for nmar_result
Tidy summary for NMAR results
Trim weights by capping and proportional redistribution
Unscale coefficients and covariance
Validate and apply scaling for engines
Validate Data for NMAR Analysis
Validate top-level nleqslv arguments and coerce invalid to defaults
Validate EL Engine Settings
Validate nmar_result
Variance-covariance for NMAR results
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>.
Useful links