Timeline for Determine rate of change in dissimilarity (distance)?
Current License: CC BY-SA 4.0
31 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| S Jun 16, 2018 at 0:26 | history | bounty ended | CommunityBot | ||
| S Jun 16, 2018 at 0:26 | history | notice removed | CommunityBot | ||
| Jun 14, 2018 at 21:52 | history | edited | theforestecologist | CC BY-SA 4.0 |
added 1 character in body
|
| Jun 14, 2018 at 1:01 | answer | added | Rob | timeline score: -1 | |
| Jun 13, 2018 at 23:43 | history | edited | theforestecologist | CC BY-SA 4.0 |
edited body
|
| Jun 11, 2018 at 23:25 | history | edited | theforestecologist | CC BY-SA 4.0 |
added 1 character in body
|
| Jun 8, 2018 at 20:20 | comment | added | user78229 | ctd... Steve Scott has several articles exploring Bayesian methods for this sites.google.com/site/stevethebayesian/… Another guy is Oded Netzer's HMM for customer segmentation ... columbia.edu/~on2110/Papers/… Finally there's proprietary software from Statistical Innovations that has ready to use tools for HMMs...the product is called Latent Gold and costs about $1,000 or so for a license. Might be worth the investment vs trying to develop a bespoke approach. | |
| Jun 8, 2018 at 20:16 | comment | added | user78229 | There's lots of ways to approach this. Aggarwal and Reddy's book Data Clustering: Algorithms and Applications covers the waterfront...it's on Amazon. If you want to retain the distance approach there are information theoretic models which use a distance function defined by Kullback-Lieber such as Brandmaier's pairwise permutation distance clustering of time series. jstatsoft.org/article/view/v067i05 describes his R module. Eamonn Keogh's SAX routines are another while many on CV recommend dynamic time warping. Finally there are hidden markov models that cluster ... ctd>> | |
| Jun 8, 2018 at 20:01 | comment | added | theforestecologist | @DJohnson Exactly!! I'm not sure whether longitudinal clustering would work or not -- your persistence suggests I didn't pick up on something when looking at it before. Could you provide a suggestion of what, specifically, you had in mind? | |
| Jun 8, 2018 at 19:44 | comment | added | user78229 | In other words you don't have an a priori definition of community and are trying to 'express' one that is dynamic and/or can change over time based on the information available from the data, is that correct? If so why isn't this an issue for longitudinal clustering? | |
| Jun 8, 2018 at 17:07 | comment | added | Isabella Ghement | What if you think of the distribution of the values stored in each distance matrix (or a relevant subset of it) and try to quantify different aspects of that distribution (e.g., percentiles)? Then you can track how specific aspects of that distribution change over time? You can focus on center and spread of the distribution and how those change over time, for instance. | |
| Jun 8, 2018 at 16:41 | history | tweeted | twitter.com/StackStats/status/1005127603190206464 | ||
| Jun 8, 2018 at 16:15 | comment | added | theforestecologist | @DJohnson, so you're right, that my most basic unit in my analysis is abundance for a species in a year of sampling in a given plot. However, as noted in my comment above, I am not sure how to express "community" other than using a dissimilarity matrix approach | |
| Jun 8, 2018 at 16:13 | comment | added | theforestecologist | So, ecologically, a community = an assemblage of interacting species. As such, I'm interested in characterizing the whole community of species per sample (plot/year). NMDS "summarizes/combines" all the species data via a distance matrix to characterize a sample (i.e., plot in a given year) as a single point. That point represents the aggregated abundance of all species in that point's given sample (plot/year). In other words, in NMDS, a point represents the whole community. The problem is, a dissimilarity matrix approach is the only way I know how to "aggregate" species abundances in this way | |
| Jun 8, 2018 at 15:21 | comment | added | user78229 | I don't understand how species rolls up into a community but it sounds like the most basic unit in your analysis is, as noted above, plot-species-year-abundance. You could modify that to include a feature that expresses community, e.g., plot-species-community-year-abundance where everything to the left of abundance is a predictor of abundance. | |
| Jun 8, 2018 at 14:44 | comment | added | theforestecologist | @DJohnson hmmm. So you're thinking that I would simply add all the species as predictors to the LGM? Or did you have something else specific in mind for how I would approach it? Or am I misunderstanding the approach you suggested? (Reminder, I have about 50-70 species in the data, with typically 20-50 species per plot. This seems to preclude adding all species to a regression model) | |
| Jun 8, 2018 at 14:30 | comment | added | user78229 | Based on my limited understanding of your data I don't see why you wouldn't be able to obtain a rate of change for a community. In other words the proposed model structure isn't limited to species level growth only. | |
| Jun 8, 2018 at 14:27 | comment | added | theforestecologist | @DJohnson Thanks for the link and suggestion. To answer your questions: 1. No, the plots are not sampled regularly or even in the same years, but each plot was sampled 12-16 times. This shouldn't matter since I know the length of time between samples and can adjust results accordingly to put into "per annum" scale. 2. I don't think the approach you mention will work for me because I am not interested in a per-species change, but rather, I'm interested in a whole community change (i.e., considering all species together). This is what drew me initially to NMDS | |
| Jun 8, 2018 at 14:13 | comment | added | user78229 | I'm not sure I completely understand the question but based on my understanding of MDS I doubt that it lends itself to an answer. First, are the 80 years of data annual and are all species measured consistently, i.e., is the panel balanced across plots, species and time? Next, why not treat it as a hierarchical and longitudinal growth model? This would involve restructuring your data matrix such that each record is plot-species-year-abundance giving about 150,000 records (37x80x50). Lots of lit on this topic, e.g., Singer's paper is a good intro ida.liu.se/~732G34/info/singer.pdf | |
| Jun 8, 2018 at 14:06 | history | edited | theforestecologist | CC BY-SA 4.0 |
deleted 1 character in body
|
| Jun 8, 2018 at 12:56 | history | edited | theforestecologist | CC BY-SA 4.0 |
edited body
|
| Jun 8, 2018 at 2:27 | history | edited | theforestecologist | CC BY-SA 4.0 |
added 310 characters in body
|
| Jun 8, 2018 at 2:21 | history | edited | theforestecologist | CC BY-SA 4.0 |
added 310 characters in body
|
| Jun 8, 2018 at 2:15 | history | edited | theforestecologist | CC BY-SA 4.0 |
added 310 characters in body
|
| Jun 8, 2018 at 2:10 | comment | added | theforestecologist |
@ttnphns As for purpose: I do not want to visualize the differences as that is what NMDS ordination allows me to do. I want to be able to determine the rate of change each sample has undergone between sampling periods. For example, I need to quantify the change in Plot 4 in 2000 vs 2010 (perhaps being represented by row 510 and 511 in my abund.data that informed the distance matrix. However, because the distance matrix represents , well, a matrix of pairwise distances bewteen all points, I'm not sure how to go about quantifying change in "distance space"
|
|
| Jun 8, 2018 at 2:07 | comment | added | theforestecologist |
@ttnphns Nope. I have a single distance matrix where each "unit" I've calculated distances for is the abundance of a given species in a given plot in a given year. The resulting distance matrix has dimensions = dim(abun.data)[1] * [(dim(abund.data)[1] - 1) / 2].
|
|
| Jun 8, 2018 at 1:51 | comment | added | ttnphns | So you have a number of distance matrices - between the species, one matrix per date - and you want to visualize the differences or trend between the matrices? Is that what you want? | |
| Jun 7, 2018 at 22:39 | history | edited | theforestecologist | CC BY-SA 4.0 |
added 353 characters in body
|
| S Jun 7, 2018 at 22:31 | history | bounty started | theforestecologist | ||
| S Jun 7, 2018 at 22:31 | history | notice added | theforestecologist | Draw attention | |
| May 8, 2018 at 3:12 | history | asked | theforestecologist | CC BY-SA 4.0 |