fix integer overflow from #151

[ecmerkle]

I encountered this problem myself and then found #151, so added the trivial fix that @avehtari describes. (but I realize the posterior package might make it irrelevant)

[codecov-commenter]

Codecov Report

Merging #155 into master will increase coverage by 0.94%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #155      +/-   ##
==========================================
+ Coverage   94.36%   95.30%   +0.94%     
==========================================
  Files          24       27       +3     
  Lines        2519     2642     +123     
==========================================
+ Hits         2377     2518     +141     
+ Misses        142      124      -18     

Impacted Files	Coverage Δ
R/effective_sample_sizes.R	97.63% <100.00%> (+13.38%)	⬆️
R/print.R	86.58% <0.00%> (-5.63%)	⬇️
R/psislw.R	87.36% <0.00%> (ø)
R/zzz.R	100.00% <0.00%> (ø)
R/extract_log_lik.R	30.00% <0.00%> (ø)
R/loo_moment_matching.R	99.20% <0.00%> (+1.58%)	⬆️
R/loo.R	98.35% <0.00%> (+5.34%)	⬆️
R/importance_sampling.R	92.55% <0.00%> (+8.51%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2eae2a6...fe50f5f. Read the comment docs.

[jgabry]

Thanks @ecmerkle! This looks good to me. @avehtari any reason we shouldn't merge this? We can still switch to calling the functions from posterior when on CRAN, but good to have this until then.

[avehtari]

It's good to merge this!