r-glennie / ctmcdive

R package to fit continuous-time Markov chain to dive data

I am a Research Fellow in Statistics at the University of St Andrews. I work on ecological problems, e.g., estimating the size of wild animal populations, quantifying how populations change over time, or finding out how animals move around.

I am actively developing software to implement my research methods, e.g. r-glennie/openpopscr and r-glennie/moveds.

Full details on my current research projects can be found at www.richardglennie.co.uk.

I can be contacted by email [email protected] or on twitter @richard_glennie if you'd like to discuss a research project. If you have any questions or issues with code in my repositories (which is without warranty) then please open an issue in the corresponding github repository.

ctmcdive's Issues

degrees of freedom calculation

Is it right? Should the random effects be taken into account too? (This didn't matter when only one smooth, with one basis size was being compared, I think, but does now multiple smooths with varying basis sizes are allowed.)

Input checking for fitting function

• check that forms is of correct length and elements are character/formula
• dat has the correct columns?

Plot/predict for intensity

At the moment, we predict/plot for duration which is more understandable, but it is useful for checking model fit to plot/predict on intensity scale. Extreme durations are not unusual for CTMC but can make the fit look poor even when the intensity is well estimated.

Add "control" exposure type and/or add generalized code for exposures

Currently only two exposure types are allowed, need to add control or the option for as many exposure types as we need.

`plot` and `predict` methods error from `example.R`

Calling these functions yields:

Error in r_dive[i] <- Ssurf[floor(mod$dat$dive[i]/dt)] : 
  replacement has length zero

make formulae more general

Currently syntax to specify the formulae is:

# setup model 
forms <- list(surface ~ 1, 
              dive ~ 1)

# fit model 
mod <- FitCTMCdive(forms, dat, model = "sd", print = TRUE)

where model takes the values: "s", "d", "sd"/"ds", or "iid".

Would like to have something more like:

# setup model 
forms <- list(surface ~ s(time, bs="ts", k=10), 
              dive ~ 1)

# fit model 
mod <- FitCTMCdive(forms, dat, print = TRUE)

installation issue on windows

Versions:

R version 4.2.0 (2022-04-22 ucrt)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 19043)

Error:

: ctmc_dive.o: too many sections (45373)
: Assembler messages:
: Fatal error: can't write 315 bytes to section .text of ctmc_dive.o: 'file too big'
: Fatal error: ctmc_dive.o: file too big

Possible references:
google/googletest#1841
https://stackoverflow.com/questions/16596876/gcc-equivalent-of-mss-bigobj

Package license

In the DESCRIPTION, the license is listed as: "MIT + License" but there is no LICENSE file present. Should this just be MIT licensed?

add EDF and k to summary

Summary for kappa

Adapt summary function so that it outputs kappa estimate. Also, print() does not seen to be working.

hierarchical structure to two exposure model

Talk to Catriona more about this to get details

conditioning on min dwell time

Need to add to calculation of indD and indS with some constant (integral start point).

AIC doesn't work for models with no/one smooths

If model does not have a smooth on both components, then AIC call fails.

Package installation on Mac

Minor issues when installing CTMCdive v0.0.2 on Mac OS v10.15.7 running R v.4.0.3:

Issues:

• CTMCdive requires Matrix v1.3.2, leading to errors related to version mismatching when another version of Matrix is installed.
• Matrix v1.3.2 appears to only be compatible with XCode Command Line Tools for Mac (v.11.5).

Fix:

• Download XCode Command Line Tools v.11.5 from https://developer.apple.com/download/all/?q=Xcode%2011.5
• Install Matrix using devtools::install_version("Matrix", version = "1.3.2", repos = "http://cran.us.r-project.org")

sessionInfo()
R version 4.0.3 (2020-10-10)
Platform: x86_64-apple-darwin17.0 (64-bit)
Running under: macOS Catalina 10.15.7

Matrix products: default
BLAS:   /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/4.0/Resources/lib/libRlapack.dylib

locale:
[1] en_GB.UTF-8/en_GB.UTF-8/en_GB.UTF-8/C/en_GB.UTF-8/en_GB.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

loaded via a namespace (and not attached):
[1] compiler_4.0.3 tools_4.0.3    yaml_2.2.1