Advanced searches left 3/3

Categorical data - PubAg

Summarized by Plex Scholar
Last Updated: 16 October 2021

* If you want to update the article please login/register

Categorical CVA biplots

Multivariate visualisation techniques such as canonical variate evaluation biplots enable for simultaneous lower-dimensional visualisation and data classification by incorporating class-specific data. With incorporating ideas from both CVA and non-linear primary element analysis biplots, a new biplot building and construction technique that improves the traditional CVA biplot by enabling categorical variables is proposed. This novel technique boosts on existing biplot building and construction in regards to classification accuracy and class separation.

Source link: https://pubag.nal.usda.gov/catalog/7410697


Partition-based feature screening for categorical data via RKHS embeddings

This paper recommends a new testing procedure for the ultrahigh dimensional data with a categorical response. By making use of the team structure among predictors, a new partition-based screening strategy is developed using the reproducing bit Hilbert space embeddings in the maximum mean discrepancy structure. Furthermore, by using the RKHS embedding, the new ranking index has a really straightforward form, and hence can be assessed quickly.

Source link: https://pubag.nal.usda.gov/catalog/7241836


The application of Local Indicators for Categorical Data (LICD) to explore spatial dependence in archaeological spaces

Local and global analyses of spatial autocorrelation are widespread in spatial archaeology. Global examinations of spatial reliance for categorical data are consistently made use of in various other fields, and neighborhood versions of these tests have lately been created. Different LICD versions have been evaluated on 2 case-studies: a historical grid, with things presence or lack videotaped for each and every cell; Historic Landscape Characterisation, with the origin of character types taped for each and every area.

Source link: https://pubag.nal.usda.gov/catalog/7221161


Conditional simulation of categorical spatial variables using Gibbs sampling of a truncated multivariate normal distribution subject to linear inequality constraints

This paper presents a technique to generate conditional categorical simulations, given an ensemble of partially conditioned categorical simulations stemmed from any kind of simulation process. The suggested conditioning technique depends on implicit functions for standing for the categorical spatial variable of rate of interest. It shows up that the recommended simulation method is a reliable approach to generate conditional categorical simulations from a set of genuine categorical simulations.

Source link: https://pubag.nal.usda.gov/catalog/7278835


Category-Adaptive Variable Screening for Ultra-High Dimensional Heterogeneous Categorical Data

In this post, we introduce a category-adaptive screening procedure with high-dimensional heterogeneous data, which is to identify category-specific important covariates. The proposition is a model-free strategy without any spec of a regression model and a flexible procedure in the feeling that the set of active variables is enabled to vary across various groups, hence making it more adaptable to suit heterogeneity. For response-selective sampling data, another major discovery of this short article is that the proposed approach works straight without any type of adjustment.

Source link: https://pubag.nal.usda.gov/catalog/6980656


MCMC for Imbalanced Categorical Data

Bayesian hierarchical versions fight data sparsity by obtaining info, while also evaluating unpredictability. However, posterior calculation offers a basic barrier to regular usage; a solitary class of formulas does not work well in all experts and settings lose time trying different types of Markov chain Monte Carlo approaches. In this program, our outcomes reveal computational complexity of Metropolis is logarithmic in example size, while data augmentation is polynomial in sample dimension. The origin of this bad efficiency of data enhancement is a disparity in between the rates at which the target thickness and MCMC step sizes concentrate. Our approaches reveal that MCMC formulas that show a comparable inconsistency will fall short in huge samples-- a result with substantial sensible effect.

Source link: https://pubag.nal.usda.gov/catalog/6713514


A nonparametric framework for inferring orders of categorical data from category-real pairs

Provided a dataset of careers and earnings, exactly how large a difference of revenues between any type of set of professions would be? Offered a dataset of traveling time records, for how long do we require to spend even more when selecting a public transport setting A rather of B to take a trip? In this paper, we recommend a framework that is able to presume orders of groups along with sizes of distinction of actual numbers between each pair of groups making use of an evaluation statistics structure. Our framework not only reports whether an order of groups exists, but it also reports sizes of distinction of each successive set of groups in the order. The outcomes of jobs getting show income inequality amongst different occupations. The stock exchange results illustrate dynamics of market dominance that can alter in time.

Source link: https://pubag.nal.usda.gov/catalog/7171573


A novel MM algorithm and the mode-sharing method in Bayesian computation for the analysis of general incomplete categorical data

In this paper, we first introduce an unique minorization-- maximization formula to compute the optimum likelihood estimates of parameters and the posterior modes for the analysis of general incomplete categorical data. Although the data enhancement formula and Gibbs sampling as the matching stochastic counterparts of the expectation-- maximization and ECM formulas are established really well, yet, little work has been done on producing stochastic variations to the existing MM algorithms. This is the first paper to propose a mode-sharing technique in Bayesian computation for general insufficient categorical data by establishing a new approval-- rejection formula helped with the proposed MM formula. The crucial suggestion is to create a course of envelope thickness indexed by a functioning criterion and to recognize a certain envelope density which can overcome the four disadvantages connected with the traditional AR formula.

Source link: https://pubag.nal.usda.gov/catalog/6461406


Source apportionment of soil heavy metals using robust spatial receptor model with categorical land-use types and RGWR-corrected in-situ FPXRF data

In-situ area portable X-ray fluorescence spectrometry is a low-cost and rapid way to enhance the example size of dirt heavy metals. Last, based on the robust spatial receptor model, this study suggested RAPCS/RGWR with categorical land-use types and RGWR-corrected in-situ FPXRF data, and its efficiency was compared with those of RAPCS/RGWR with none or one kind of auxiliary data. Outcomes showed that the efficiencies of the adjustment models for in-situ FPXRF data remained in the order of RGWR > GWR > MLR, and the relative enhancement of RGWR over MLR varied from 52. 6% to 70. 71% for each and every HM; categorical land-use types dramatically influenced the concentrations of soil Zn, Cu, As, and Pb; the greatest estimation precision for resource contributions was acquired by RAPCS/RGWR _ LU&FPXRF, and the lowest estimate accuracy was obtained by basic RAPCS/RGWR.

Source link: https://pubag.nal.usda.gov/catalog/7199969


A new correlation coefficient between categorical, ordinal and interval variables with Pearson characteristics

A prescription exists for a useful and new correlation coefficient, ϕK, based on a number of refinements to Pearson's hypothesis test of freedom of two variables. Mainly, it works constantly in between categorical, interval and ordinal variables, essentially by treating each variable as categorical, and can consequently be used to compute correlations between variables of blended type. The strength of ϕK resembles Pearson's relationship coefficient, and is equal in case of a bivariate normal input distribution.

Source link: https://pubag.nal.usda.gov/catalog/7002954

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions