scholarly journals On Selection Criteria for the Tuning Parameter in Robust Divergence

Entropy ◽  
2021 ◽  
Vol 23 (9) ◽  
pp. 1147
Author(s):  
Shonosuke Sugasawa ◽  
Shouto Yonekura

Although robust divergence, such as density power divergence and γ-divergence, is helpful for robust statistical inference in the presence of outliers, the tuning parameter that controls the degree of robustness is chosen in a rule-of-thumb, which may lead to an inefficient inference. We here propose a selection criterion based on an asymptotic approximation of the Hyvarinen score applied to an unnormalized model defined by robust divergence. The proposed selection criterion only requires first and second-order partial derivatives of an assumed density function with respect to observations, which can be easily computed regardless of the number of parameters. We demonstrate the usefulness of the proposed method via numerical studies using normal distributions and regularized linear regression.

Entropy ◽  
2020 ◽  
Vol 22 (3) ◽  
pp. 270
Author(s):  
Elena Castilla ◽  
Nirian Martín ◽  
Leandro Pardo ◽  
Konstantinos Zografos

This paper presents a model selection criterion in a composite likelihood framework based on density power divergence measures and in the composite minimum density power divergence estimators, which depends on an tuning parameter α . After introducing such a criterion, some asymptotic properties are established. We present a simulation study and two numerical examples in order to point out the robustness properties of the introduced model selection criterion.


Agronomy ◽  
2020 ◽  
Vol 11 (1) ◽  
pp. 57
Author(s):  
Nesrine Chrigui ◽  
Duygu Sari ◽  
Hatice Sari ◽  
Tuba Eker ◽  
Mehmet Fatih Cengiz ◽  
...  

The chickpea leafminer, Liriomyza cicerina (Rondani), is one of the most destructive insect pests of cultivated chickpea (Cicer arietinum L.) in the Mediterranean region under field conditions. For sustainable and environmentally friendly chickpea production, efforts have been devoted to managing the leafminer via decreasing the use of insecticides. Breeding of new resistant varieties is not only an efficient and practical approach, but also cost-effective and environmentally sensitive. To improve resistant varieties, breeders need reliable biochemical selection criteria that can be used in breeding programs. The first objective was to investigate the possible introgression of resistance to the leafminer from C. reticulatum Ladiz. (resistant) to C. arietinum (susceptible), then, to estimate inheritance of resistance to the leafminer for efficient breeding strategies, and finally, to study organic acid contents as selection criteria. Recombinant inbred lines (RILs) and their parents were evaluated using a visual scale of 1–9 (1 = free from leafminer damage and 9 = mines in more than 91% of the leaflets and defoliation greater than 31%) in the field under natural infestation conditions after the susceptible parent and check had scores of >7 on the visual scale. Superior RILs were found for resistance to the leafminer, and agro-morphological traits indicating that introgression of resistance to leaf miner from C. reticulatum to C. arietinum could be possible using interspecific crosses. The inheritance pattern of resistance to the leafminer in RILs was shown to be quantitative. Organic acids, including oxalic, malic, quinic, tartaric, citric and succinic acids in RILs grown in the field under insect epidemic conditions and in the greenhouse under non-infested conditions were detected by using high performance liquid chromatography (HPLC). In general, organic acids were found to be higher in resistant RILs than susceptible RILs. Path and correlation coefficients showed that succinic acid exhibited the highest direct effects on resistance to the leafminer. Multivariate analyses, including path, correlation and factor analyses suggested that a high level of succinic acid could be used as a potential biochemical selection criterion for resistance to leafminer in chickpea. Resistant RILs with a high seed yield resembling kabuli chickpea can be grown directly in the target environments under leaf miner infestation conditions.


1989 ◽  
Vol 26 (03) ◽  
pp. 512-523
Author(s):  
Clifton Sutton

Codes having all pairs of words separated by a Hamming distance of at least d are stochastically constructed by sequentially packing randomly generated q-ary n-tuples. Estimates of the random packing densities are obtained by repeated simulation. Using non-linear regression to fit the estimated densities, an asymptotic approximation formula is obtained for the packing densities which depends only on q, n, d, and an empirical constant.


2021 ◽  
Vol 2021 ◽  
pp. 1-13
Author(s):  
Jiulun Fan ◽  
Jipeng Yang

Circular histogram represents the statistical distribution of circular data; the H component histogram of HSI color model is a typical example of the circular histogram. When using H component to segment color image, a feasible way is to transform the circular histogram into a linear histogram, and then, the mature gray image thresholding methods are used on the linear histogram to select the threshold value. Thus, the reasonable selection of the breakpoint on circular histogram to linearize the circular histogram is the key. In this paper, based on the angles mean on circular histogram and the line mean on linear histogram, a simple breakpoint selection criterion is proposed, and the suitable range of this method is analyzed. Compared with the existing breakpoint selection criteria based on Lorenz curve and cumulative distribution entropy, the proposed method has the advantages of simple expression and less calculation and does not depend on the direction of rotation.


2020 ◽  
Vol 18 (2) ◽  
pp. 58-73
Author(s):  
Halimin Herjanto ◽  
◽  
Alexandra Chilicki ◽  
Chidchanok Anantamongkolkul ◽  
Erin McGuinness ◽  
...  

Consumers use online e-reviews as a popular tool for information and obtaining guidance. E-reviews have therefore become an important barometer in conducting product evaluations, and more importantly, to make purchasing decisions. This includes decisions about hotel selection. For hospitality industry marketers, the information in e-reviews is particularly important in translating and understanding consumer-specific needs. The current study brings valuable awareness to the limited academic research into hotel selection criteria among solo-traveling females. TripAdvisor’s top 25 list of cost-efficient hotels worldwide received 345 total consumer reviews. Noteworthy findings of the current study show unique selection criteria considered important to the solo traveling female including a hotel-provided cell phone programmed with emergency local contact information, and a nearby or on-premise automated teller machine. Study results also indicate that stewardship service, such as intimate and personalized hotel staff who “go the extra mile” is an important selection criterion. Also discussed are research limitations and implications.


Entropy ◽  
2020 ◽  
Vol 22 (4) ◽  
pp. 399 ◽  
Author(s):  
Marco Riani ◽  
Anthony C. Atkinson ◽  
Aldo Corbellini ◽  
Domenico Perrotta

Minimum density power divergence estimation provides a general framework for robust statistics, depending on a parameter α , which determines the robustness properties of the method. The usual estimation method is numerical minimization of the power divergence. The paper considers the special case of linear regression. We developed an alternative estimation procedure using the methods of S-estimation. The rho function so obtained is proportional to one minus a suitably scaled normal density raised to the power α . We used the theory of S-estimation to determine the asymptotic efficiency and breakdown point for this new form of S-estimation. Two sets of comparisons were made. In one, S power divergence is compared with other S-estimators using four distinct rho functions. Plots of efficiency against breakdown point show that the properties of S power divergence are close to those of Tukey’s biweight. The second set of comparisons is between S power divergence estimation and numerical minimization. Monitoring these two procedures in terms of breakdown point shows that the numerical minimization yields a procedure with larger robust residuals and a lower empirical breakdown point, thus providing an estimate of α leading to more efficient parameter estimates.


Sign in / Sign up

Export Citation Format

Share Document