Dear all, I have spent some time reading most of the very interesting and helpful texts you suggested (thanks again!) and played with sample landmark data. Unfortunately, I did not understand all of the math, so I still fell like I am missing some important points and misunderstand some details. Therefore, I would be very grateful for any feedback on my thoughts and the conclusions I drew so far.
(I calculated all the Procrustes distances with the function “riemdist” of the R package "shapes" as it seemed to me that this was an implementation of the equations in Kendall 1984. If I below talk about tangent space projection, I mean the projection performed by the geomorph function “gpagen”; I suppose that this function orthogonally projects the GPA-superimposed shapes into a space tangent to the (hyper-)hemisphere with radius 1, not into a space tangent to Kendall’s shape space.) >From the literature and tests with sample triangles, I would say that, if landmark sets are superimposed by GPA such that they are located on the hemisphere of radius 1, the following statements are true for triangles: 1. The arclength between two landmark configurations on the hemisphere equals the Procrustes distance between the corresponding points (measured as arc length) in Kendall’s shape space (i.e. sphere of radius 0.5) if (i) one of the landmark configurations is the pole/reference (mean shape) on the hemisphere, (ii) if both landmark configurations are located on the same meridian in Kendall’s shape space, or (iii) if both landmark configurations are located on a circle formed by two opposing meridians in Kendall’s shape space as long as the distance is measured across the upper pole. The Euclidean distance between two landmark configurations on the hemisphere that satisfy any of the conditions (i)-(iii) is shorter than the arclength (= Procrustes distance) on the hemisphere. Assuming an orthogonal projection, the corresponding Euclidean distance in tangent space is shorter than the Euclidean distance “on” the hemisphere, except for the case that two landmark configurations are additionally located on the same line of latitude. 2. For landmark configurations on the same line of latitude on the hemisphere: There are cases for which the Euclidean distance between two landmark configurations is longer than the Procrustes distance between the corresponding points in Kendall’s shape space (e.g. if the landmark configurations are located on meridians that meet at an angle of 90°). In other cases, the Euclidean distance “on” the hemisphere is shorter than the Procrustes distance (e.g. if the two landmark configurations on the same line of latitude additionally satisfy condition (iii) from 1.). An orthogonal tangent space projection does not alter the Euclidean distance between two landmark configurations on the same line of latitude. 3. If two landmark configurations neither have the same latitude nor the same longitude either the Euclidean distance “on” the hemisphere or in tangent space might be the better approximation for the Procrustes distance. There is a higher linear correlation between the Procrustes distance and the Euclidean distance “on” the hemisphere than between the Procrustes distance and the Euclidean distance in tangent space. I also compared the different distance measures for 100 random triangles and random landmark configurations with arbitrary numbers and dimensions of landmarks. In all cases there were Euclidean distances “on” the (hyper-)hemisphere and in tangent space that were shorter than the corresponding Procrustes distance and such that were longer. There always was a higher linear correlation between the Procrustes distance and the Euclidean distance “on” the (hyper-)hemisphere than between the Procrustes distance and the Euclidean distance in tangent space. Considering these observations, I would expect that the Euclidean distance “on” the (hyper-)hemisphere is the better approximation for the Procrustes distance. To get a feeling for the effect of a tangent space projection on permutational MANOVAs as they are performed in numerous papers, I generated random shapes in groups, superimposed them with the gpagen function without and with tangent space projection, and tested for significant group differences using procD.lm from geomorph. Using the coordinates on the hemisphere without tangent space projection resulted in the correct decision on significance slightly more often than using the coordinates in tangent space if the differences between groups were small. For large or no differences between groups, I noticed no differences in the number of correct decisions. In order to better understand the reasons for this observation, I tried to calculate the F values following Anderson (2001 Austal Ecology) using the Procrustes distance, the Euclidean distance “on” the hemisphere, and the Euclidean distance in tangent space to generate different distance matrices; F values were calculated for the original and permuted groups. I am absolutely not sure if this was a valid approach because the relation illustrated in Fig. 2 in Anderson (2001) might not be true for the Procrustes distance (Anderson stated that the relation was true for any distance measure but I miss a proof), yet the results seem reasonable to me. If this was a valid approach and if the F values calculated from the Procrustes distances were assumed to be the correct ones (and if I implemented everything correctly), then the F values calculated based on the Euclidean distance “on” the hyper-hemisphere seem to be the better approximation (almost always higher correlation with F values calculated from Procrustes distances). This somehow surprises me, because, as far as I understand the calculations of the within- and among-group sum of squares (SS) from the distance matrix in Anderson (2001), the calculation of SS values involves something like an implicit calculation of (group) mean “shapes”. If Euclidean distances were used, these mean “shapes” would be the arithmetic mean coordinates of groups or of all landmark configurations. At least in the case of using Euclidean distances between the shapes on the hyper-hemisphere, the centroid size of the mean “shape” would be smaller than 1. This would mean that, in the case of using the Euclidean distances between the shapes on the hyper-hemisphere, mean “shapes” are used that are no shapes by definition and lie inside the hyper-hemisphere. On the other hand: is there really a problem if one used the Euclidean distances between points in the Euclidean space in which the hyper-hemisphere is embedded and allowed mean “shapes” to be located inside the hyper-hemisphere just for the purpose of approximating the Procrustes distances in Kendall’s shape space? Considering the observations above, I am still asking myself what the advantage of a tangent space projection is if the coordinates are used to measure distances or in permutational MANOVAs that are based on Euclidean distances. There, for sure, are sets of landmark configurations for which a tangent space projection partly corrects the distortions in the distances between landmark configurations on the hyper-hemisphere (if compared to the distances in Kendall’s shape space). In other cases, I would expect that a tangent space projection distorted the Euclidean distances even more (e.g. shortening already too short Euclidean distances between landmark sets on the same meridian of the hyper-hemisphere and not shortening already too long Euclidean distances between certain landmark sets with the same latitude). Some last thoughts that have not become clear to me from the literature: Does a GPA onto the sample mean shape result in the optimal superimposition of any pair of landmark configurations in the sample (specifically with regard to rotation)? From the observations above I would guess the answer was no. If the answer was no, I wonder if this explained why, when performing a PCA of GPA-aligned landmark configurations, there is one more non-zero eigenvalue than there are dimensions in Kendall’s shape space (given that the sample sizes was large enough)? The tangent space has the same dimension like the shape space, but is this loss of one dimension compared to the GPA-aligned data related to a better superimposition of the landmark configurations or is it because a curved surface is flattened? Thanks a lot for any comments, corrections, or further insights! Best wishes, Karo PS: Attached you can find the scripts I based my observations on in case anyone is interested in taking a closer look. I hope that the scripts are self-explanatory; if they are not, I am happy to provide more information. [email protected] schrieb am Sonntag, 12. September 2021 um 13:51:47 UTC+2: > Dear all, > > Many thanks for the additional explanations and literature > recommendations!!! > > I unfortunately still did not have the time to read all of the recommended > texts, so I still might miss important points. But from what I read so far > and especially from the additional enlightening explanations of Prof. Rohlf > I realized that my understanding of the hemisphere with the radius of 1 was > wrong. I think I generalized the illustrations that show the Procrustes > distance (great circle distance on hemisphere) between the reference and a > triangle on the hemisphere too much by assuming that the great circle > distance on the hemisphere equaled the Procrustes distance in Kendall’s > shape space for all possible pairs of triangle shapes; I somehow did not > note if this equivalence was restricted to specific cases in the > accompanying texts. I will pay special attention to this during my > readings, and also to how the distances are “distorted” during tangent > space projection and what this potentially means for MANOVAs. > > All the best, > Karolin > > PS1: Please forget about the script > “Compare_distance_measures_in_original_and_tangent_space.R” I sent earlier; > my calculation of the Procrustes distance in there is wrong, because I did > not understand the hemisphere correctly. I am sorry for that! > > PS2: In case someone has similar questions in mind as I have: One > additional text that I think has not been mentioned so far but that helped > me is Slice (2001) Landmark coordinates aligned by Procrustes analysis do > not lie in Kendall's shape space, Systematic Biology. > > Am Fr., 10. Sept. 2021 um 00:06 Uhr schrieb Leandro Rabello Monteiro < > [email protected]>: > >> Also, if you have a mathematics background, I would recommend Goodall >> (1991) (https://doi.org/10.1111/j.2517-6161.1991.tb01825.x), besides the >> papers from Kendall in the 1980s. I have colleagues that are mathematicians >> that were able to understand what it was all about in a couple of days >> after reading this material. >> Leandro >> >> Em quinta-feira, 9 de setembro de 2021 às 10:36:19 UTC-3, >> [email protected] escreveu: >> >>> I was a little vague there. Unfortunately there are several relevant >>> distances. Usually best to think of Procrustes distances in terms of angles >>> or great circle distances. Max between two shapes is pi/2. Distances in >>> the tangent space are linear but with a maximum of 2 for shapes with >>> projections maximally far apart - but misleading because they are actually >>> the same shape (Procrustes distance of almost zero). Points around the >>> outer limits of the hemisphere have a Euclidean distance of 1 to the >>> reference triangle but a Procrustes distance of pi/2 = 1.57. Problem of >>> trying to project surface of a sphere onto a flat map. I shouldn't have >>> tried to simplify so much. Looking at pictures may be safer than words. >>> >>> It is less confusing for the range found in most biological data where >>> shape variation is fortunately usually "small". >>> >>> >>> *F. James Rohlf * >>> Distinguished Professor, Emeritus and Research Professor >>> Depts: Anthropology and Ecology & Evolution >>> Stony Brook University >>> >>> On 9/9/2021 5:36:16 AM, Karolin Engelkes <[email protected]> wrote: >>> Thank you all so much for the literature recommendations and, Prof. >>> Rohlf, for the explanations! Especially the part on the PCA and the >>> patterns of distribution in the different spaces is very helpful for me, >>> but also gives me a lot to think about. So please give me some time to go >>> through all of it and through the literature and digest it. >>> >>> @ Prof. Rohlf: One question immediately comes to my mind that is related >>> to your statement that the “distance in the tangent space will be a little >>> larger [than the distances on the hemisphere] due to the projection“. Does >>> this refer to an orthogonal or a stereographic projection? I would expect >>> the opposite for an orthogonal projection (at least from taking a ruler and >>> measuring the linear distances between the “reference” [intersection of the >>> hemisphere with the y-axis] and, respectively, the points B, C, and D in >>> your Fig. 4 in the paper from 1999). >>> >>> Thanks again and best wishes, >>> Karo >>> >>> Am Do., 9. Sept. 2021 um 09:54 Uhr schrieb mahendiran mylswamy < >>> [email protected]>: >>> >>>> Another nice piece of writing 'on Shape' Theory by Prof. Mac leod, pdf >>>> attached, may help to understand and appreciate the concepts well. >>>> >>>> On Thu, Sep 9, 2021 at 7:35 AM [email protected] < >>>> [email protected]> wrote: >>>> >>>>> Good questions. The topic can be confusing and difficult to visualize >>>>> - especially for 3D landmark data. >>>>> >>>>> The 1999 paper in the Journal of Classification that Adams mentions >>>>> was my attempt to describe its practical relevance in morphometrics. My >>>>> 1999 paper in Hystrix may also be of interest to you. You may wish to >>>>> play >>>>> with my tpsTri software also as it was used for some of the figures in >>>>> those papers. >>>>> >>>>> Easiest to at first just think of variation in shapes of triangles in >>>>> 2 dimensions. As you mention, Kendall's shape space for triangles can be >>>>> visualized as the surface of a sphere of radius 1/2. What convinced >>>>> me of the importance of Kendall's shape space was that Kendall showed >>>>> that >>>>> the distribution of all possible triangles was a uniform distribution in >>>>> Kendall's shape space. After a GPA the distribution of shapes is on >>>>> a hemisphere of radius 1 (corresponding to centroid size of 1). I am not >>>>> sure if anyone has given a good name for this hemisphere corresponding to >>>>> all possible triangles Procrustes aligned to any single shape. In 1999 I >>>>> called it a "preshape space of triangles aligned to a reference >>>>> triangle". >>>>> Not very snappy! Perhaps I should have called it something like the >>>>> "Slice >>>>> hemisphere" as Dennis Slice first showed it to me and was puzzled why it >>>>> was not a surface of a sphere. >>>>> >>>>> The distribution of triangles is not, however, uniform on the surface >>>>> of the hemisphere. As you mention, conventional multivariate methods >>>>> assume >>>>> linear spaces and use linear matrix algebra. The tangent space is as you >>>>> mention the projection of points from the surface of the hemisphere onto >>>>> a >>>>> plane that passes just through the point on the hemisphere that >>>>> corresponds >>>>> to the reference triangle (not really a "reference" just the mean shape >>>>> in >>>>> practical applications). I have called it Kendall's tangent space but I >>>>> probably should have named it after John Kent as he influenced my >>>>> understanding. Something I found fascinating was that the distribution of >>>>> all possible triangles was again uniform in the projection within the >>>>> circular distribution (Kendall showed that also). The importance, to me, >>>>> of >>>>> being uniform was that if I see some pattern in the distribution >>>>> (clusters, >>>>> covaniance, etc.) then it implies something about the distribution of >>>>> shapes not just an artifact of the mathematical operations used to create >>>>> the projection (as in the case of EDMA and some other earlier >>>>> statistical >>>>> approaches suggested for analyzing shape variation). >>>>> >>>>> Of course, for very small variation in shape the points will be close >>>>> to their mean so that distances on the surface of the hemisphere (thus >>>>> close to their projections) and distances in the tangent space will be >>>>> very >>>>> similar (though distance in the tangent space will be a little larger due >>>>> to the projection). Might be good enough for some studies but if one does >>>>> not do the projection then you will find that a PCA of your GPA aligned >>>>> data (assuming large n > 2p-4) will not yield 4 zero eigenvalues as it >>>>> should with centering, rotation, and size removed. Only 3 will be zero >>>>> (i.e., computationally numbers like 10^-15 or so). The 4th smallest might >>>>> be "only" 10^-8 or so. That is a result of the curved shape of the >>>>> hemisphere. If you do the projection then the last 4 eigenvalues will be >>>>> essentially zero as the curvature is now gone. >>>>> >>>>> An alternative is to perform the multivariate analysis directly in >>>>> Kendall's shape space. Kent (I cannot locate the references right now but >>>>> it was in early 2000s) showed one could, for example, perform a >>>>> generalization of a PCA directly in the curved surface. Some odd >>>>> properties >>>>> as eigenvectors were great circles on the curved surfaces as I remember). >>>>> >>>>> One can generalize, of course from triangles to shapes with more >>>>> landmarks and to landmarks in 3 dimensions. The 3-dimensional case is >>>>> more >>>>> complicated to try to visualize because the simplest case requires 5 >>>>> dimensions to represent not just 3 so one cannot just look at the space. >>>>> It >>>>> also has some more complicated properties. You could look at the book >>>>> "Shape & shape theory" by Kendall, Barden, and Le (1999). It presents a >>>>> way >>>>> to visualize variation in the 5-dimensional shape space. Was not an easy >>>>> read for me but perhaps I should try again! >>>>> >>>>> This distinction between shape space and tangent space is not of much >>>>> importance in practical applications where biological variation tends to >>>>> be >>>>> small compared to all possible variation among p landmarks and because >>>>> one >>>>> usually only looks at the distribution along the first few eigenvectors >>>>> with the largest eigenvalues but I prefer to have computations match what >>>>> one expects theoretically rather than just being good approximations. >>>>> When >>>>> programming being just "close enough" could hide subtle bugs. Getting rid >>>>> of that known artifact also allows one to try to possibly interpret those >>>>> eigenvectors with the smallest eigenvalues as they correspond to the most >>>>> stable aspects of possible shape variation (i.e., least varying due to >>>>> development, environment, etc.). >>>>> >>>>> Does this help or confuse more? >>>>> >>>>> *F. James Rohlf * >>>>> Distinguished Professor, Emeritus and Research Professor >>>>> Depts: Anthropology and Ecology & Evolution >>>>> Stony Brook University >>>>> >>>>> On 9/8/2021 2:26:38 PM, Adams, Dean [EEOB] <[email protected]> wrote: >>>>> >>>>> Karolin, >>>>> >>>>> >>>>> >>>>> A reading of Rohlf 1999 may help. >>>>> >>>>> >>>>> >>>>> Dean >>>>> >>>>> >>>>> >>>>> Rohlf, F.J. 1999. Shape statistics: Procrustes superimpositions and >>>>> tangent spaces. Journal of Classification. 16:197-223. >>>>> >>>>> >>>>> >>>>> Dr. Dean C. Adams >>>>> >>>>> Distinguished Professor of Evolutionary Biology >>>>> >>>>> Director of Graduate Education, EEB Program >>>>> >>>>> Department of Ecology, Evolution, and Organismal Biology >>>>> >>>>> Iowa State University >>>>> >>>>> https://faculty.sites.iastate.edu/dcadams/ >>>>> >>>>> phone: 515-294-3834 <(515)%20294-3834> >>>>> >>>>> >>>>> >>>>> *From:* [email protected] <[email protected]> *On >>>>> Behalf Of *[email protected] >>>>> *Sent:* Tuesday, September 7, 2021 7:04 AM >>>>> *To:* Morphmet <[email protected]> >>>>> *Subject:* [MORPHMET2] Questions about Kendall’s shape space and >>>>> tangent space projection >>>>> >>>>> >>>>> >>>>> Dear Morphometricians, >>>>> >>>>> I am currently trying to understand the mathematical backgrounds of >>>>> landmark-based geometric morphometrics. Some questions arose that we >>>>> could >>>>> not answer during discussions in our lab which is why I hope you can help >>>>> - >>>>> many thanks in advance! >>>>> >>>>> The first question is: What exactly is “Kendall’s shape space”? If I >>>>> understand Kendall’s (1984) statement in Eq. 4 correctly, the shape space >>>>> is a quotient space; the elements are equivalence classes of pre-shapes >>>>> (a >>>>> fiber on the pre-shape sphere becomes one element in shape space). The >>>>> elements of the equivalence classes have less “coordinates” (vector >>>>> elements) than the original landmark configuration and lie on a >>>>> hyperdimensional sphere with a radius of 1. In Theorem 2 Kendall (1984) >>>>> states that the shape space for triangles is isometric to a >>>>> three-dimensional sphere with a radius of 0.5. The triangles on this >>>>> sphere >>>>> with a radius of 0.5 are represented by three Cartesian coordinates that >>>>> are calculated from the original landmark configuration (Kendall 1984, >>>>> section 5), whereas the triangles are represented by equivalence classes >>>>> in >>>>> shape space. >>>>> In several publications I now find illustrations of a hemisphere of >>>>> radius 1 and a sphere of radius 0.5 (both share one point at the pole); >>>>> those publications usually use the full landmark set. The sphere of >>>>> radius >>>>> 0.5 is often termed “Kendall’s shape space” (sometimes with a reference >>>>> to >>>>> triangles, sometimes not). So, how does this fit with the definitions and >>>>> statements in Kendall (1984)? Is there a publication that extends Kendall >>>>> (1984) to the use of full landmark configurations and explains how they >>>>> are >>>>> (mathematically) related to the sphere with radius 0.5 (for all numbers >>>>> and >>>>> dimensions of landmarks)? Related to this question: what do the points on >>>>> the sphere of radius 0.5 in those publications look like? Are they >>>>> equivalence classes, full landmark configurations, or 3 cartesian >>>>> coordinates representing triangles? Are they really scaled to unit >>>>> centroid >>>>> size as the shapes on the pre-shape sphere [= elements of equivalence >>>>> classes in shape space]? >>>>> >>>>> The second question is: Why do we need a tangent space projection? I >>>>> understand that the superimposed landmark configurations lie on a >>>>> hyper-hemisphere and I know the argument that standard statistical >>>>> procedures need a linear space. Yet, the superimposed landmark >>>>> configurations are matrices or vectors, depending on how they are >>>>> formatted, for which we can compute Euclidean distances. Where exactly do >>>>> the statistical tests go wrong if we use the superimposed landmark >>>>> configurations without tangent space projection and calculate Euclidean >>>>> distances? >>>>> If I, for example, think about MANOVAs as suggested by Anderson (2001, >>>>> Austral Ecology), I guess that the mean shapes of the groups need to be >>>>> calculated to be able to calculate the different sums of squares. If the >>>>> mean “shape” is calculated by group-wisely simply calculating the mean of >>>>> each of the coordinates, the resulting mean “shape” of each group lies >>>>> within the hyper-hemisphere of radius 1. So the mean “shape” is not a >>>>> shape >>>>> because the centroid size is not standardized. Yet, if I got all distance >>>>> calculations correctly (see attached R-script >>>>> “Compare_distance_measures_in_original_and_tangent_space.R”), I find that >>>>> the Euclidean distances between the mean “shapes” inside the >>>>> hyper-hemisphere are slightly closer to the corresponding Procrustes >>>>> distances than the Euclidean distances in tangent space; the Procrustes >>>>> distances have been calculated by rescaling the mean “shapes” to unit >>>>> centroid size followed by determining the arc length between them. If the >>>>> mean “shapes” inside the hyper-hemisphere are rescaled to unit centroid >>>>> size, then the Euclidean distances between them are even closer to the >>>>> Procrustes distances. >>>>> In addition, if I simulate groups of landmark configurations, >>>>> superimpose them without and with tangent space projection, and test for >>>>> significant differences between the groups, I feel that the decision on >>>>> the >>>>> significance of the group differences is correct slightly more often if >>>>> it >>>>> is based on the superimposed landmark sets without tangent space >>>>> projection >>>>> (not exhaustively or formally tested; see R-script >>>>> “compare_ProcrustesMANOVA_in_original_and_tangent_space.R”). >>>>> >>>>> And one last, more general question: If all landmark configurations >>>>> are superimposed onto a common mean shape, does this also minimize the >>>>> Procrustes distances (measured as arc length) between all pairs of >>>>> landmark >>>>> configurations and between the mean shapes of sub-groups of landmark >>>>> configurations? >>>>> >>>>> Thanks a lot for your insights! >>>>> >>>>> Kind regards >>>>> Karo >>>>> -- >>>>> Dr. Karolin Engelkes >>>>> Institute of Evolutionary Biology and Animal Ecology >>>>> University of Bonn >>>>> >>>>> Phone: +49 (0) 228 73 5481 <+49%20228%20735481> >>>>> >>>>> An der Immenburg 1 >>>>> 53121 Bonn >>>>> Germany >>>>> >>>>> -- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Morphmet" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> To view this discussion on the web visit >>>>> https://groups.google.com/d/msgid/morphmet2/ac177d3d-b28a-41b2-bfb6-a7a9f8cd73d0n%40googlegroups.com >>>>> >>>>> <https://groups.google.com/d/msgid/morphmet2/ac177d3d-b28a-41b2-bfb6-a7a9f8cd73d0n%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>> . >>>>> >>>>> -- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "Morphmet" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to [email protected]. >>>>> To view this discussion on the web visit >>>>> https://groups.google.com/d/msgid/morphmet2/Mailbird-a92a6485-db7d-4c94-82a2-397208a82e0f%40stonybrook.edu >>>>> >>>>> <https://groups.google.com/d/msgid/morphmet2/Mailbird-a92a6485-db7d-4c94-82a2-397208a82e0f%40stonybrook.edu?utm_medium=email&utm_source=footer> >>>>> . >>>>> >>>> >>>> >>>> -- >>>> *************************************** >>>> M Mahendiran, Ph D >>>> Sr. Scientist - Division of Wetland Ecology >>>> Salim Ali Centre for Ornithology and Natural History (SACON) >>>> Anaikatti (PO), Coimbatore - 641108, TamilNadu, India >>>> Tel: 0422-2203100 (Ext. 122), 2203122 (Direct), Mob: 09787320901 >>>> Fax: 0422-2657088 >>>> http://www.sacon.in/staff/dr-m-mahendiran/ >>>> >>>> P Please consider the environment before printing this email >>>> >>> -- >> > You received this message because you are subscribed to a topic in the >> Google Groups "Morphmet" group. >> To unsubscribe from this topic, visit >> https://groups.google.com/d/topic/morphmet2/ex8LLCDZkpo/unsubscribe. >> To unsubscribe from this group and all its topics, send an email to >> [email protected]. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/morphmet2/2a1d6878-29de-46fc-a5eb-711cdefbd81cn%40googlegroups.com >> >> <https://groups.google.com/d/msgid/morphmet2/2a1d6878-29de-46fc-a5eb-711cdefbd81cn%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> > -- You received this message because you are subscribed to the Google Groups "Morphmet" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/morphmet2/a0f849f1-db23-4d0c-a828-116119ef927bn%40googlegroups.com.
compare_ProcrustesMANOVA_in_original_and_tangent_space.R
Description: Binary data
_compare_F-values_derived_by_different_distance_calculations.R
Description: Binary data
_compare_different_distance_measures_for_specific_triangles.R
Description: Binary data
_compare_different_distance_measures_for_random_shapes.R
Description: Binary data
