Translation should comment about the future

[jamesrobertlloyd]

e.g. SE * Per - the posterior mean tends to zero but this is just due to uncertainty about the period

In general, SE*Per should talk about a range of plausible periods

[jamesrobertlloyd]

SE*Per now talks about credible periods - should be possible to calculate how quickly predictive amplitude decays

[jamesrobertlloyd]

Will also discuss some simple model checks

[jamesrobertlloyd]

Now talking about extrapolation and model checks - to be worked on continuously from now on