GIN 2016-20

[QLLZ]

Data pull: HASH: 6772ea9da4f43ddc3da11c494a573af2ca3dd3c8 config

[QLLZ]

Model run: HASH: 6772ea9da4f43ddc3da11c494a573af2ca3dd3c8

[QLLZ]

Country data report

[eclee25]

Mild sd_w and rho traceplot issues and many Rhats exceeding threshold. However, there are almost no cases reported in this country and the maps and fits look otherwise okay.

Opinion: Second opinion requested but leaning towards Approve

[javierps]

The number of cases here is very low (6 at national level), so don't expect to run well.

Suggestion: Reject and run with no-w model.

[eclee25]

Opinion: Temp Approve and rerun with spatial_effect: no

[QLLZ]

Config

Data pull & Model run:

HASH: 211913e0eb7d570ed62c48bc1ae9ecd6671a6cbc

[QLLZ]

country data report

[QLLZ]

Poor convergency due to low number of cases. Modeled cases look good.

It seems that this was not moved to the no-w column, thus this was run with w.

Suggestion: rerun with spatial-effect: no.

[QLLZ]

Updated no-w config

[QLLZ]

Updated no-w country data report

[QLLZ]

Results look better.

Suggestion: accept.

[eclee25]

Very few overall cases in every year and therefore limited subnational data. Opinion: Approve the no-W model

[eclee25]

Try no-mixture sd_w prior model for minimal subnational data

[QLLZ]

Data pull, model run:

HASH: 05f598b95b5d42ac0049963f8a0b7e04d6f1dbd7

[javierps]

Rho convergence issues, std_w and ws OK, all Rhats OK. Fig 26 shows some disagreement between chains, but numbers are very small.

Suggestion: accept.

[eclee25]

Accept