Skip to main content
rw2's user avatar
rw2's user avatar
rw2's user avatar
rw2
  • Member for 8 years, 5 months
  • Last seen this week
2 votes
2 answers
3k views

Why is confidence width narrower, when sample proportions are smaller?

1 vote
1 answer
2k views

R glm - should my random effect be a fixed effect?

3 votes
1 answer
1k views

Predicting a realistic distribution using regression

2 votes
1 answer
1k views

What family of GLM when response is Bray-Curtis dissimilarity? Should I use "adonis" instead?

1 vote
2 answers
967 views

Fitting binomial regression model in R - correct formula, significance testing, and over-dispersion

1 vote
1 answer
410 views

Logistic regression - approaches for showing probabilities for all combinations of predictors

0 votes
0 answers
337 views

How to test for differences in gradient between time-series

2 votes
1 answer
335 views

Exciting visualisation for "observed value" vs "expected range"

4 votes
1 answer
305 views

How to account for spatial auto-correlation when testing for differences in community composition

1 vote
1 answer
285 views

Can KD-tree introduce bias to nearest neighbor search?

1 vote
0 answers
263 views

Accounting for spatial structure in constrained ordination analysis (vegan)

2 votes
1 answer
219 views

Model overall effect of predictor within categories

3 votes
1 answer
124 views

Is it possible to test for strongest predictor, if all variables are identical within a random effect grouping?

1 vote
0 answers
124 views

Obtaining individual level classifications from predicted probabilities

4 votes
1 answer
90 views

Is there a way to optimise the spread of explanatory variables in experimental design?

1 vote
0 answers
71 views

Methods for incorporating a feature in a predictive model that is identical for many (but not all) instances

1 vote
1 answer
68 views

Which proportions are particularly high or low - compare confidence intervals, or logistic regression?

0 votes
0 answers
60 views

Paired sign test, when expected difference is not 50:50

3 votes
1 answer
56 views

Should I apply my treatment as a continuous gradient, or in blocks?

0 votes
0 answers
41 views

Is there a method for evaluating predictive models that accounts for difference between predicted/actual distributions?

0 votes
0 answers
30 views

Classification model - metrics to optimise so that number of Predicted positives equals the number of Actual positives