scholarly journals On the Estimated Variances of Regression Coefficients in Misspecified Error Components Models

1991 ◽  
Vol 7 (3) ◽  
pp. 369-384 ◽  
Author(s):  
Philippe J. Deschamps
Keyword(s):  
Variance Components ◽  
Asymptotic Variance ◽  
Matrix Multiplication ◽  
Regression Coefficients ◽  
Linear Combinations ◽  
Error Components ◽  
Restricted Model ◽  
Time Component ◽  
Error Components Models ◽  
The Difference

In a regression model with an arbitrary number of error components, the covariance matrix of the disturbances has three equivalent representations as linear combinations of matrices. Furthermore, this property is invariant with respect to powers, matrix addition, and matrix multiplication. This result is applied to the derivation and interpretation of the inconsistency of the estimated coefficient variances when the error components structure is improperly restricted. This inconsistency is defined as the difference between the asymptotic variance obtained when the restricted model is correctly specified, and the asymptotic variance obtained when the restricted model is incorrectly specified; when some error components are improperly omitted, and the remaining variance components are consistently estimated, it is always negative. In the case where the time component is improperly omitted from the two-way model, we show that the difference between the true and estimated coefficient variances is of order greater than N–1 in probability.

scholarly journals Subgraph Network Random Effects Error Components Models: Specification and Testing

2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Gabriel Montes-Rojas
Keyword(s):  
Random Effects ◽  
Quadratic Forms ◽  
Variance Components ◽  
Network Effects ◽  
Interbank Market ◽  
Error Components ◽  
Finite Samples ◽  
Random Components ◽  
Error Components Models ◽  
Lagrange Multiplier Tests

Abstract This paper develops a subgraph random effects error components model for network data linear regression where the unit of observation is the node. In particular, it allows for link and triangle specific components, which serve as a basal model for modeling network effects. It then evaluates the potential effects of ignoring network effects in the estimation of the coefficients’ variance-covariance matrix. It also proposes consistent estimators of the variance components using quadratic forms and Lagrange Multiplier tests for evaluating the appropriate model of random components in networks. Monte Carlo simulations show that the tests have good performance in finite samples. It applies the proposed tests to the Call interbank market in Argentina.

Confidence intervals on linear combinations of variance components that are unrestricted in sign

1990 ◽  
Vol 35 (3-4) ◽  
pp. 135-143 ◽  
Author(s):  
Naitee Ting ◽  
Richard K. Burdick ◽  
Franklin A. Graybill ◽  
S. Jeyaratnam ◽  
Tai-Fang C. Lu
Keyword(s):  
Confidence Intervals ◽  
Variance Components ◽  
Linear Combinations

A note on autoregressive error components models

1985 ◽  
Vol 28 (2) ◽  
pp. 231-245 ◽  
Author(s):  
P. Sevestre ◽  
A. Trognon
Keyword(s):  
Error Components ◽  
Error Components Models

scholarly journals PENERAPAN METODE MULTI ATTRIBUTE UTILITY THEORY (MAUT) DALAM PEMETAAN TINGKAT DAMPAK BENCANA BANJIR DI KABUPATEN BANTUL

Telematika ◽  
2020 ◽  
Vol 17 (1) ◽  
pp. 26
Author(s):  
Afif Irfan Abdurrahman ◽  
Bambang Yuwono ◽  
Yuli Fauziah
Keyword(s):  
Utility Theory ◽  
Matrix Multiplication ◽  
Flood Disaster ◽  
Accuracy Rate ◽  
Affected Area ◽  
A Value ◽  
Multi Attribute Utility Theory ◽  
The Matrix ◽  
The Difference ◽  
The Impact

Flood disaster is a dangerous disaster, an event that occurs due to overflow of water resulting in submerged land is called a flood disaster. Almost every year Bantul Regency is affected by floods due to high rainfall. The flood disaster that struck in Bantul Regency made the Bantul District Disaster Management Agency (BPBD) difficult to handle so that it needed a mapping of the level of the impact of the flood disaster to minimize the occurrence of floods and provide information to the public.This study will create a system to map the level of impact of floods in Bantul Regency with a decision support method namely Multi Attribute Utility Theory (MAUT). The MAUT method stage in determining the level of impact of flood disasters through the process of normalization and matrix multiplication. The method helps in determining the areas affected by floods, by managing the Indonesian Disaster Information Data (DIBI). The data managed is data on criteria for the death toll, lost victims, damage to houses, damage to public facilities, and damage to roads. Each criteria data has a value that can be used to determine the level of impact of a flood disaster. The stages for determining the level of impact of a disaster require a weighting calculation process. The results of the weighting process display the scoring value which has a value of 1 = low, 2 = moderate, 3 = high. To assist in determining the affected areas using the matrix normalization and multiplication process the process is the application of the Multi Attribute Utility Theory (MAUT) method.This study resulted in a mapping of the level of impact displayed on google maps. The map view shows the affected area points and the level of impact of the flood disaster in Bantul Regency. The mapping produced from the DIBI data in 2017 produced the highest affected area in the Imogiri sub-district. The results of testing the data can be concluded that the results of this study have an accuracy rate of 95% when compared with the results of the mapping previously carried out by BPBD Bantul Regency. The difference in the level of accuracy is because the criteria data used are not the same as the criteria data used by BPBD in Bantul Regency so that the accuracy rate is 95%.

scholarly journals Relevance of reproductive correlates in response of diamondback moth (Lepidoptera: Plutellidae) to plant quality

2013 ◽  
Vol 45 (2) ◽  
pp. 10 ◽  
Author(s):  
M. Soufbaf ◽  
Y. Fathipour ◽  
M.P. Zalucki ◽  
J. Karimzadeh
Keyword(s):  
Diamondback Moth ◽  
Reproductive Potential ◽  
Fertility Rate ◽  
Regression Coefficients ◽  
Initial Population ◽  
Reproductive Parameters ◽  
Plant Nitrogen ◽  
Fecundity Rate ◽  
Standardized Regression Coefficients ◽  
The Difference

To study the relationships between leaf nitrogen and the reproductive potential of diamondback moth, all reproductive parameters of this pest raised on two canola cultivars were evaluated. A standardized regression coefficient (<em>&beta;</em>) was used as an index for nitrogen-reproduction relationship strength. The only difference between net fecundity rate and net fertility rate is <em>h<sub>x</sub></em>&rsquo;s effect, but the difference in their standardized regression coefficients was not significant [<em>&beta;</em>=+0.934 (R<sup>2</sup>=0.87, F<sub>1,4</sub>=27.34, P=0.006) and <em>&beta;</em>=+0.922 (R<sup>2</sup>=0.85, F<sub>1,4</sub>=22.825, P=0.009)]. Accordingly, gross fecundity rate and gross fertility rate differ only in <em>h<sub>x</sub></em>&rsquo;s effect, but the difference in standardized regression coefficients again was not significant [<em>&beta;</em>=0.895 (R<sup>2</sup>=0.8, F<sub>1,4</sub>=16.159, P=0.016)-0.890 (R<sup>2</sup>=0.79, F<sub>1,4</sub>=15.266, P=0.017)=0.005]. As gross fecundity rate differs from net fecundity rate only in midpoint survivorship (<em>L<sub>x</sub></em>)&rsquo;s effect, it is understood that survivorship could affect the plant nitrogen&ndash;fecundity relation considerably (standardized coefficients difference=0.044) and could be a critical parameter in insectplant interactions. But, the terms of reproductive parameters, <em>i.e. L<sub>x</sub> </em>and <em>h<sub>x</sub></em>, showed the same effect on the strength of nitrogen-fecundity regression statistically, even though <em>L<sub>x</sub></em> has been selected frequently by many researchers as an important fitness correlate. Measuring the hatch rate could be recommended in trophic interactions studies due to its being easier to apply, more robust, and quicker to accomplish than measurement of survivorship; however, it is important as an indicator in combination with brood size for determining the initial population size of an insect herbivore.

scholarly journals Difference in Bias Approach for Commutability Assessment: Application to Frozen Pools of Human Serum Measured by 8 Direct Methods for HDL and LDL Cholesterol

2015 ◽  
Vol 61 (8) ◽  
pp. 1107-1113 ◽  
Author(s):  
William J Korzun ◽  
Göran Nilsson ◽  
Lorin M Bachmann ◽  
Gary L Myers ◽  
Ikunosuke Sakurabayashi ◽  
...  
Keyword(s):  
Random Error ◽  
Ldl Cholesterol ◽  
Direct Method ◽  
Direct Methods ◽  
Measurement Procedure ◽  
Critical Value ◽  
Reference Measurement ◽  
Reference Procedure ◽  
Error Components ◽  
The Difference

Abstract BACKGROUND We used a difference in bias approach to evaluate the commutability of 4 frozen serum pools for 8 direct methods for measurement of HDL and LDL cholesterol (HDLC and LDLC). METHODS Freshly collected nonfrozen sera from 138 diseased and 37 nondiseased patients and 4 frozen pools from the CDC Lipid Standardization Program were measured by direct methods and by the beta-quantification reference measurement procedure of the CDC. We used an error components model to estimate the difference in the bias component of error plus its uncertainty for frozen pools vs patient samples between the direct method and the reference procedure. Frozen pools with bias differences less than a critical value determined by either medical requirements for bias or the random error components of the measurement procedures were considered commutable. RESULTS On the basis of medical requirement criteria, 1 of the 4 frozen pools was commutable for most of the HDLC methods for both diseased and nondiseased patients, and none was commutable for LDLC methods. On the basis of random error criteria, all of the frozen pools were generally commutable for all of the HDLC methods for both diseased and nondiseased patients, and 1 of the 4 frozen pools was generally commutable for most of the LDLC methods for both diseased and nondiseased patients. CONCLUSIONS Commutability was assessed as the closeness of agreement of the difference in bias between a reference material and a set of patient samples. Criteria for commutability could be based on fixed medical requirements for bias or on random error components.

Sign in / Sign up

Export Citation Format

Share Document