Model convergence investigation log =================================== StatMedAnalysis | Original downloadable worksheet | 6 November 2026 Resolve optimization warnings before interpreting fitted parameters. How to use this worksheet Record the exact warning, model specification, software version and optimizer settings. Check data scaling, sparse patterns, identifiability and starting values. Compare objective values and parameter stability across justified runs. Preserve unsuccessful attempts and avoid interpreting a returned coefficient table as proof of convergence. Save a working copy and keep the original blank version. Use public, hypothetical or aggregate information only; do not enter patient identifiers, credentials or confidential records. Keep source evidence in your approved project workspace. Mark unresolved text fields UNKNOWN or NOT REPORTED with a reason; never substitute zero for an unknown count. Project/version __________________ Prepared by role/date __________________ Working fields 1. Model family and formula Response/evidence: _________________________________________________________ ___________________________________________________________________________ 2. Warning and stopping criterion Response/evidence: _________________________________________________________ ___________________________________________________________________________ 3. Optimizer and tolerances Response/evidence: _________________________________________________________ ___________________________________________________________________________ 4. Scaling and sparse-data checks Response/evidence: _________________________________________________________ ___________________________________________________________________________ 5. Starting values and alternative runs Response/evidence: _________________________________________________________ ___________________________________________________________________________ 6. Objective and gradient evidence Response/evidence: _________________________________________________________ ___________________________________________________________________________ 7. Parameter/SE stability Response/evidence: _________________________________________________________ ___________________________________________________________________________ 8. Resolution or unresolved status Response/evidence: _________________________________________________________ ___________________________________________________________________________ Hypothetical worked case A fictional mixed model returns coefficients with a boundary variance warning. Increasing iteration count alone may not resolve an unsupported random-effect structure; examine identifiability and the substantive model. Decision rule and limitations Do not report unstable estimates as a successful fit. Explain unresolved convergence or boundary issues and use a defensible simpler or alternative model when necessary. Decision _______________________ Remaining uncertainty ____________________ Next action ____________________ Owner / review date ______________________ Independent check and discrepancy resolution _______________________________ Official method or reporting reference checked 6 November 2026 https://www.statsmodels.org/stable/pitfalls.html This is an original planning aid, not an official checklist, validated instrument, clinical recommendation or certificate of expert approval. Apply the referenced guidance to the specific design and record any additional requirements.