Function reference • silentema

Package overview

silentema-package silentema: silentema: dynamic missingness graphs and sensitivity analysis for EMA data

Step 1. Declare and check: dynamic missingness graphs

Build a dm-graph from the motif taxonomy, query it, and read the recoverability report.

dm_graph(): Declare a dynamic missingness graph (dm-graph)

plot(<dm_graph>): Plot a dm-graph

dsep(): d-separation in a dm-graph

recoverability(): Recoverability report for a dm-graph

Step 2. Test: was silence informative?

silence_test(): The silence test: does the state after a skipped prompt differ?

sensor_gap_test(): The sensor-gap test: does an always-observed sensor differ at skipped prompts?

fatigue_check(): Descriptive check for burden or fatigue in the response sequence

Step 3. Estimate the two-level VAR(1)

make_pairs(): Build adjacent prompt pairs from long experience-sampling data

fit_pairs(): Within-person VAR(1) estimated from complete adjacent pairs

summary(<pairs_fit>) coef(<pairs_fit>) vcov(<pairs_fit>) confint(<pairs_fit>) nobs(<pairs_fit>): Standard errors and confidence intervals for the lagged coefficients

fit_ipw(): Inverse-probability-weighted within-person VAR(1) for observed context confounding

fit_fiml(): Full-information maximum likelihood under missing at random (state-space EM)

loglik_fiml(): Log-likelihood of the two-level VAR(1) at given parameters (MAR)

Step 4. Sensitivity analysis for self-censoring

fit_tilt(): Tilted (self-censoring-adjusted) within-person VAR(1) at a fixed value of the sensitivity parameter

tilt_profile(): Sensitivity profile over a grid of self-censoring values

plot(<tilt_profile>): Plot a tilt profile

break_even(): Break-even sensitivity values and identified sets from a tilt profile

bounds_support(): Worst-case (support) bounds for person means under arbitrary nonresponse

Step 5. Calibrate the sensitivity parameter

calibrate_delta(): Calibrate the self-censoring sensitivity parameter from a design feature

plot(<delta_calibration>): Plot the model-implied curve of a calibration

simulate_from_fit(): Simulate data from a fitted tilt model

Step 6. Report

missingness_declaration(): Missingness declaration for preregistrations and reports

Simulation and population parameters

simulate_ema(): Simulate experience-sampling data with a declared missingness mechanism

default_params(): Default population parameters used in the simulation studies of Yu (2026)

stationary_cov(): Stationary covariance of a VAR(1) process

partial_cors(): Partial correlations from a covariance matrix