John Best

icon indicating a website link https://www.posdefecology.net icon indicating an email [email protected]

icon company Washington Department of Fish & Wildlife icon location Olympia, WA icon location Operations Research Specialist at WDFW (hydropower and salmon), PhD student in Quantitative Ecology and Resource Management at the UW (spatiotemporal models).

Results 4 issues of John Best

No pivoting for CHOLMOD factors

1 comment

`whiten!` and `unwhiten` use [`chol_lower` to extract the lower triangular component of Cholesky decompositions](https://github.com/JuliaStats/PDMats.jl/blob/9fe6b3cb4f5cb256405940a42773d4eddde25d1a/src/chol.jl#L7). For `SuiteSparse.CHOLMOD.Factor{T}`, this extracts the lower triangular factor, but does not account for pivoting. This results...

Change-of-support references

Following up on today's change-of-support discussion, I found [this thread](https://groups.google.com/forum/#!msg/r-inla-discussion-group/SL5McDqCRyI/j1veDTtJaDQJ;context-place=forum/r-inla-discussion-group) on the INLA mailing list. It mentions that you can do a change-of-support (integrating over an area) with INLA by...

Building better meshes

A key part of getting valid approximations to a spatial field in INLA is forming a "good" mesh (triangulation). This section of the "SPDE Tutorial" on the INLA website describes...

Some models (e.g. VAST) fail to compile with newer compilers

17 comment

## Issue I am having trouble getting newer versions of VAST to compile. Others are encountering similar difficulties (e.g. James-Thorson-NOAA/VAST#140). In that issue Jim suggested posting here for more feedback...

External