Reputation
12,846
Top tag
Next privilege 15,000 Rep.
Protect questions
Badges
2 36 69
Newest
 Nice Answer
Impact
~640k people reached

May
12
answered logistic regression in r outside range [0,1]
May
12
revised Comparing Euclidean distances
added 25 characters in body; edited title
May
11
awarded  Nice Answer
May
8
revised Different behaviour for local regression function
no need for tags in titles
May
7
revised Best Subset Selection Questions
added 2 characters in body
May
7
revised Best Subset Selection Questions
needed some = signs
May
7
comment Chi-Square transformation on a partially unknown matrix
Re 2. Right; so it sounds like it would be OK to just consider the selection of OTUs you do have as the "community" you wish to analyse, and take the row/col/total sum from the data (relativised if that is what you are using - i.e apply the entire transformation to the data you are using)
May
7
answered Best Subset Selection Questions
May
7
comment Chi-Square transformation on a partially unknown matrix
As for the other question; I'd work out first what your OTUs represent, can you consider them the community of interest? As a limnologist, I might go an look at all species of zooplankton I can find in my samples, but what about the algae or other inverts, etc? I can rightly consider the zooplankton in isolation for the purposes of the analysis. Can you consider your OTUs that you have data for in that or a similar way?
May
7
comment Chi-Square transformation on a partially unknown matrix
CCA is not redundant on those data; the two square root terms are important weightings & even the analysis itself is a weighted linear model which is why you can't do a chi-square transformation followed by RDA and get exactly the same results as CCA. The RDA version is an unweighted analysis.
May
7
comment Doubt with a distance based Redundancy analysis
@StasK Bray-Curtis is the name of a distance/dissimilarity metric: en.wikipedia.org/wiki/Bray%E2%80%93Curtis_dissimilarity This is the distance matrix that is to be embedded in principal coordinates & then ordinated using dbRDA. It's popular in ecology for measuring dissimilarity between samples/assemblages. In RDA, the abundance is measured as a set of $m$ orthogonal linear combinations of the predictor variables. It all starts to get messy when transformations & the distance-based version is being used.
May
7
comment Doubt with a distance based Redundancy analysis
@StasK dbRDA is distance-based RDA and RDA is Redundancy Analysis (reduced rank regression sensu Rao), essentially a reduced-rank multivariate multiple regression. In dbRDA, one embeds a distance matrix in a Euclidean space using principal coordinates analysis, and then take the principal coordinates and use them as the response matrix in in the RDA.
May
6
answered Chi-Square transformation on a partially unknown matrix
May
6
comment Hellinger transformation with relative data
From your comment in your other related Q it makes me wonder if you have the real row totals or just the row totals of the subset of columns (species) that your are working with. I don't think it would change my answer much if you don't have the actual row totals, but just note that if this is the case, then you are not doing a Hellinger transformation on the full data set but relativising to the subset when doing the Hellinger.
May
6
comment Doubt with a distance based Redundancy analysis
You question needs more context and detail to explain what it is you are showing, where it came from, and what the setting is. Not everyone (in fact few people) here will be familiar with dbRDA or even multivariate analyses of thius type.
May
6
answered Doubt with a distance based Redundancy analysis
May
6
comment Doubt with a distance based Redundancy analysis
I suspect you are referring to a particular implementation of the method, but this is actually a more general question related to constrained ordination.
May
6
answered Post-hoc test after a GLM with binary data
May
6
comment Post-hoc test after a GLM with binary data
"I now want to run a post-hoc test to see which levels (Cem1, Cem2, or Cem3) are significant" significant in what sense? Do you want to test pairwise equality of group means?
May
6
revised Hellinger transformation with relative data
deleted 3 characters in body