Entropy maximization principle and selection of the order of an autoregressive Gaussian process

1978 ◽  
Vol 30 (2) ◽  
pp. 263-270 ◽  
Author(s):  
Ryoichi Shimizu
Keyword(s):  
Gaussian Process ◽  
Entropy Maximization ◽  
Entropy Maximization Principle ◽  
Selection Of ◽  
Maximization Principle

scholarly journals Selection of Features on Mining Techniques for Classification

2018 ◽  
Vol 7 (S1) ◽  
pp. 108-111
Author(s):  
Gurrampally Kumar ◽  
S. Mohan ◽  
G. Prabakaran
Keyword(s):  
Feature Selection ◽  
Linear Regression ◽  
Gaussian Process ◽  
Statistical Method ◽  
Correlation Coefficient ◽  
Euclidean Distance ◽  
Experimental Results ◽  
Selection Of

Feature selection has been developed by several mining techniques for classification. Some existing approaches couldn’t remove the irrelevant data from dataset for class. Thus it needs the selection of appropriate features that emphasize its role in classification. For this it consider the statistical method like correlation coefficient to identify the features from feature set whose data are very important for existing classes. The several methods such as Gaussian process, linear regression and Euclidean distance have taken into consideration for clarity of classification. The experimental results reveal that the proposed method identifies the exact relevant features for several classes.

scholarly journals Simulation of a Gaussian stationary process with a stable correlation function with a given reliability and accuracy

2020 ◽  
pp. 89-95
Author(s):  
M. Yu. Petranova
Keyword(s):  
Correlation Function ◽  
Gaussian Process ◽  
Stationary Process ◽  
Rates Of Convergence ◽  
Random Processes ◽  
Random Series ◽  
Gaussian Stationary Process ◽  
Specific Process ◽  
Selection Of

In this paper, the representation of random processes in the form of random series with uncorrelated members obtained in the work by Yu. V. Kozachenko, I.V. Rozora, E.V. Turchina (2007) [1]. Similar constructions were studied in the book by Yu. V. Kozachenko and others. [2] in the general case. However, there are additional difficulties in construction of models of specific process, such as, for example, selection of the appropriate basis in L_2(R). In this paper, models are constructed that approximate the Gaussian process with a stable correlation function $\rho_{\alpha} (h) = E X_{\alpha}(t + h) X_{\alpha}(t) = B^2 \exp{-d|h|^{\alpha}}, \alpha > 0, d > 0$ with parameter $\alpha = 2$, which is a centered stationary process with a given reliability and accuracy in the space L_p ([0,T]). And also the rates of convergence of the models are found, the corresponding theorems are formulated. Methods of representation and main properties of the process with a stable correlation function $\rho_2(h) = B^2 \exp{-d|h|^2}, d > 0$ are considered. As a basis in the space L_2(T) Hermitian functions are used.

scholarly journals Elastic Tie-Pointing—Transferring Chronologies between Records via a Gaussian Process

Radiocarbon ◽  
2013 ◽  
Vol 55 (4) ◽  
pp. 1975-1997 ◽  
Author(s):  
Timothy J Heaton ◽  
Edouard Bard ◽  
Konrad A Hughen
Keyword(s):  
Monte Carlo ◽  
Gaussian Process ◽  
Monte Carlo Methods ◽  
Full Account ◽  
Information Selection ◽  
General Methodology ◽  
Credible Intervals ◽  
Selection Of ◽  
Independent Model ◽  
Age Estimates

We consider a general methodology for the transferral of chronologies from a master reference record containing direct dating information to an undated record of interest that does not. Transferral is achieved through the identification, by an expert, of a series of tie-points within both records that are believed to correspond to approximately contemporaneous events. Through tying of the 2 records together at these points, the reference chronology is elastically deformed onto the undated record. The method consists of 3 steps: creation of an age-depth model for the reference record using its direct dating information; selection of the tie-points and translation of their age estimates from the reference to the undated record; and finally, creation of an age-depth model for the undated record using these uncertain tie-point age estimates. Our method takes full account of the uncertainties involved in all stages of the process to create a final chronology within the undated record that allows joint age estimates to be found together with their credible intervals. To achieve computational practicality, we employ a Gaussian process to create our age-depth models. Calculations can then be performed exactly without resort to extremely slow Monte Carlo methods involving multiple independent model fits that would be required by other age-depth models.

Sign in / Sign up

Export Citation Format

Share Document