Investing across periods with Mahalanobis distances

2017 ◽  
Author(s):  
Edouard Senechal ◽  
Brian Singer
Keyword(s):  
Mahalanobis Distances

scholarly journals Morphometric Analysis of the Host Effect on Phenotypical Variation ofBelminus ferroae(Hemiptera: Triatominae)

2015 ◽  
Vol 2015 ◽  
pp. 1-12 ◽  
Author(s):  
Claudia Magaly Sandoval Ramirez ◽  
Elsa Evelia Nieves Blanco ◽  
Reinaldo Gutiérrez Marin ◽  
Diego Alexander Jaimes Mendez ◽  
Nelcy Ortiz Rodríguez ◽  
...  
Keyword(s):  
Morphometric Analysis ◽  
Sexual Size Dimorphism ◽  
Etiologic Agent ◽  
Wing Shape ◽  
Centroid Size ◽  
Mahalanobis Distances ◽  
Size Dimorphism ◽  
Host Effect ◽  
Successive Generations ◽  
Greater Wing

The Triatominae subfamily includes hematophagous insects, well known for their role as vectors for theTrypanosoma cruziparasite, etiologic agent of Chagas’ disease.Belminus ferroaeis a triatomine that showed an increased demographic fitness when cockroaches were used as hosts. Here we compare the centroid size (CS) and wing shape betweenB. ferroaeparents and three successive generations (O1, O2, and O3) of their offspring fed on cockroaches or mice under laboratory conditions. Morphometric analysis of the wings bugs fed on cockroaches showed a significant reduction in CS in both sexes among all generations. Sexual size dimorphism (SSD) was observed in the insects fed on cockroaches (O2 and O3), as well as those bugs fed on mice (O2). Differences in the shape of wings were observed between parental and offspring wings when fed on mice, but not in males (O1, O2, and O3) or females (O1 and O2) fed on cockroaches. There was a greater wing shape similarity between the cockroach-fed offspring and their parents according to the Mahalanobis distances. Our results support the idea of higher adaptation of this Triatominae with arthropod hosts.

scholarly journals To the question of the origin of the cultural assortment   of Dagestan apricots through assessing the variability   of leaf morphological characteristics

2019 ◽  
Vol 14 (3) ◽  
pp. 17-24
Author(s):  
D. M. Anatov ◽  
Z. M. Аsadulaev ◽  
R. M. Osmanov ◽  
K. I. Akhmedova
Keyword(s):  
Leaf Shape ◽  
Morphological Characteristics ◽  
Principal Component ◽  
Leaf Size ◽  
Shape Index ◽  
Botanical Garden ◽  
Mahalanobis Distances ◽  
Traditional Cultivars ◽  
Academy Of Sciences ◽  
Degree Of Similarity

Aim. The paper presents the results of assessment of the indigenous nature and  degree of similarity of apricot cultivars growing in the collection of the Mountain  Botanical Garden, Gunib, Dagestan, Russia based on a comparative analysis of the  variability of leaf morphological characteristics.   Material and Methods. The material assessed consisted of 33 apricot cultivars of  various ecological and geographical origins aggregated in the following groups: (a)  Dagestan – traditional cultivars; (b) Moscow ‐ selection from the Tsytsin Main Moscow Botanical Garden, Russian Academy of Sciences based on wild forms of Tajikistan and Kyrgyzstan; (c) European and (d) Asian ‐ from Central Asia, Tajikistan, China  and Altai.   Results. The closeness of Dagestan and European varieties in comparison with Asian  and Moscow varieties was shown. Most Dagestan (16 of 19) and European varieties  have round‐shaped leaves (leaf shape index 80‐ 100%), while those from Asia and  the Moscow Botanical Garden have leaves which are elongated elliptical and oval  (60‐80%). Using the method of principal component analysis (PCA), it was established that most cultivars of Dagestan origin have similar leaf shapes and sizes, of  which Tlama kurak (wide‐round), Hekobarsh (elongated) were distinguished by leaf  shape and Esdelik by leaf size.   Conclusion. Based on a discriminant analysis (Squared Mahalanobis Distances), it  was found that the indices of indicators of leaf attributes (width/length of leaf lamina; petiole length/length of lamina; apex angle/corner of leaf base) are more reliable criteria for differentiating apricot varieties into ecological and geographical  groups than their morphological characteristics.  

scholarly journals Pseudotext Injection and Advance Filtering of Low-Resource Corpus for Neural Machine Translation

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Michael Adjeisah ◽  
Guohua Liu ◽  
Douglas Omwenga Nyabuga ◽  
Richard Nuetey Nortey ◽  
Jinling Song
Keyword(s):  
Machine Translation ◽  
Language Processing ◽  
Training Data ◽  
Target Language ◽  
Similarity Metrics ◽  
Mahalanobis Distances ◽  
Parallel Corpora ◽  
Parallel Corpus ◽  
Low Resource ◽  
Sentence Level

Scaling natural language processing (NLP) to low-resourced languages to improve machine translation (MT) performance remains enigmatic. This research contributes to the domain on a low-resource English-Twi translation based on filtered synthetic-parallel corpora. It is often perplexing to learn and understand what a good-quality corpus looks like in low-resource conditions, mainly where the target corpus is the only sample text of the parallel language. To improve the MT performance in such low-resource language pairs, we propose to expand the training data by injecting synthetic-parallel corpus obtained by translating a monolingual corpus from the target language based on bootstrapping with different parameter settings. Furthermore, we performed unsupervised measurements on each sentence pair engaging squared Mahalanobis distances, a filtering technique that predicts sentence parallelism. Additionally, we extensively use three different sentence-level similarity metrics after round-trip translation. Experimental results on a diverse amount of available parallel corpus demonstrate that injecting pseudoparallel corpus and extensive filtering with sentence-level similarity metrics significantly improves the original out-of-the-box MT systems for low-resource language pairs. Compared with existing improvements on the same original framework under the same structure, our approach exhibits tremendous developments in BLEU and TER scores.

scholarly journals Interpopulational variability of Pinus sylvestris L. in eight Polish localities expressed in morphological and anatomical traits of needles

2014 ◽  
Vol 63 (1) ◽  
pp. 67-76 ◽  
Author(s):  
Maria A. Bobowicz ◽  
Adolf F. Korczyk
Keyword(s):  
Pinus Sylvestris ◽  
Minimum Spanning Tree ◽  
Multivariate Statistical ◽  
Mahalanobis Distances ◽  
The North ◽  
Flat Side ◽  
Standing Trees ◽  
Statistical Hypotheses ◽  
Resin Canals ◽  
Interpopulational Variability

Two-year old needles were collected from 272 standing trees of <i>Pinus sylvestris</i> L., representing 8 Polish populations. The needles were studied in respect to IS morphological and anatomical traits. The obtained data were subjected to multivariate statistical analysis in an attempt to delineate interpopulational variability. Multivariate analysis of variance with testing of statistical hypotheses and discriminant analysis were conducted. Mahalanobis distances were calculated between each of population in pairs and their significance was estimated using Hotelling T<sup>2</sup> statistics. On the basis of the shortest Mahalanobis distances a minimum spanning tree was constructed and on the basis of Euklidean distances hierarchy grouping was performed. A large majority of the populations was found to differ significantly from the remaining populations. The population from Bolewice proved to be most divergent. The principal variables which proved capable of discriminating between populations were found to include: needle length, the number of stomata on the flat side of the needle and the number of resin canals. Using Bryant's test, the studied populations were found to belong to two geographic groups: the North-Polish one or the South-Polish one.

scholarly journals Monitoring Robust Estimates for Compositional Data

2021 ◽  
Vol 50 (2) ◽  
pp. 16-37
Author(s):  
Valentin Todorov
Keyword(s):  
Compositional Data ◽  
Real Life ◽  
R Package ◽  
Diagnostic Tools ◽  
Data Set ◽  
Mahalanobis Distances ◽  
Robust Estimates ◽  
Real Life Data ◽  
Manufactured Exports ◽  
Parameter Values

In a number of recent articles Riani, Cerioli, Atkinson and others advocate the technique of monitoring robust estimates computed over a range of key parameter values. Through this approach the diagnostic tools of choice can be tuned in such a way that highly robust estimators which are as efficient as possible are obtained. This approach is applicable to various robust multivariate estimates like S- and MM-estimates, MVE and MCD as well as to the Forward Search in whichmonitoring is part of the robust method. Key tool for detection of multivariate outliers and for monitoring of robust estimates is the Mahalanobis distances and statistics related to these distances. However, the results obtained with thistool in case of compositional data might be unrealistic since compositional data contain relative rather than absolute information and need to be transformed to the usual Euclidean geometry before the standard statistical tools can be applied. Various data transformations of compositional data have been introduced in the literature and theoretical results on the equivalence of the additive, the centered, and the isometric logratio transformation in the context of outlier identification exist. To illustrate the problem of monitoring compositional data and to demonstrate the usefulness of monitoring in this case we start with a simple example and then analyze a real life data set presenting the technologicalstructure of manufactured exports. The analysis is conducted with the R package fsdaR, which makes the analytical and graphical tools provided in the MATLAB FSDA library available for R users.

FUZZY C-MEAN ALGORITHM BASED ON MAHALANOBIS DISTANCES AND BETTER INITIAL VALUES

2007 ◽  
pp. 1398-1404 ◽  
Author(s):  
HSIANG-CHUAN LIU ◽  
JENG-MING YIH ◽  
SHIN-WU LIU
Keyword(s):  
Mahalanobis Distances ◽  
Initial Values

The Cheilosia canicularis group (Diptera: Syrphidae): species delimitation and evolutionary relationships based on wing geometric morphometrics

Zootaxa ◽  
2008 ◽  
Vol 1825 (1) ◽  
pp. 40 ◽  
Author(s):  
JASMINA LUDOŠKI ◽  
LJUBINKA FRANCUSKI ◽  
ANTE VUJIĆ ◽  
VESNA MILANKOV
Keyword(s):  
Sequence Data ◽  
Balkan Peninsula ◽  
Wing Shape ◽  
Canonical Variate ◽  
Wing Size ◽  
Shape Variation ◽  
Geometric Morphometric ◽  
Mahalanobis Distances ◽  
Size And Shape ◽  
Upgma Cluster Analysis

A landmark-based geometric morphometric approach was used to assess differences in the size and shape of wing among/within three species of the Cheilosia canicularis group (Diptera: Syrphidae): C. canicularis, C. himantopus and C. orthotricha. Wing size and shape variation was observed from 25, 176 and 41 specimens of C. canicularis, C. himantopus and C. orthotricha, respectively, collected from six localities on the Balkan Peninsula. Significant differences in wing size were obtained among the analysed species and canonical variate analysis showed that wing shape was sufficiently different to allow the correct classification of 73% individuals of C. canicularis, 80% of C. orthotricha and 94% of C. himantopus, and clear delimitation of the species pairs C. canicularis/C. orthotricha and C. himantopus/C. orthotricha. In all analysed species, the consistent sex dimorphism in wing shape was observed indicating that female specimens had shorter and broader wings than males. The UPGMA cluster analysis based on squared Mahalanobis distances revealed close accordance with previously published phylogenetic relationships of these species indicated by allozyme and DNA sequence data analysis. Our results suggested that wing parameters contain useful information in quantification phenotypic variation and identification of species in this challenging group for taxonomy and systematics.

Missing data imputation over academic records of electrical engineering students

2019 ◽  
Vol 28 (4) ◽  
pp. 487-501 ◽  
Author(s):  
Esteban Jove ◽  
Patricia Blanco-Rodríguez ◽  
José-Luis Casteleiro-Roca ◽  
Héctor Quintián ◽  
Francisco Javier Moreno Arboleda ◽  
...  
Keyword(s):  
Missing Data ◽  
Engineering Students ◽  
Missing Values ◽  
Multivariate Adaptive Regression Splines ◽  
Missing Data Imputation ◽  
Mahalanobis Distances ◽  
General Terms ◽  
Academic Records ◽  
Adaptive Regression ◽  
Performance And Evaluation

Abstract Nowadays, the quality standards of higher education institutions pay special attention to the performance and evaluation of the students. Then, having a complete academic record of each student, such as number of attempts, average grade and so on, plays a key role. In this context, the existence of missing data, which can happen for different reasons, leads to affect adversely interesting future analysis. Therefore, the use of imputation techniques is presented as a helpful tool to estimate the value of missing data. This work deals with the academic records of engineering students, in which imputation techniques are applied. More specifically, it is assessed and compared to the performance of the multivariate imputation by chained equations methodology, the adaptive assignation algorithm (AAA) based on multivariate adaptive regression splines and a hybridization based on self-organisation maps with Mahalanobis distances and AAA algorithm. The results show that proposed methods obtain successfully results regardless the number of missing values, in general terms.

Sign in / Sign up

Export Citation Format

Share Document