scholarly journals Validation of single‐step GBLUP genomic predictions from threshold models using the linear regression method: An application in chicken mortality

2020 ◽  
Vol 138 (1) ◽  
pp. 4-13
Author(s):  
Matias Bermann ◽  
Andres Legarra ◽  
Mary Kate Hollifield ◽  
Yutaka Masuda ◽  
Daniela Lourenco ◽  
...  
Keyword(s):  
Linear Regression ◽  
Single Step ◽  
Regression Method ◽  
Linear Regression Method ◽  
Threshold Models ◽  
Genomic Predictions

scholarly journals PSXII-37 Validation of single-step GBLUP genomic predictions from threshold models using the linear regression method: an application in chicken mortality

2020 ◽  
Vol 98 (Supplement_4) ◽  
pp. 246-247
Author(s):  
Matias Bermann ◽  
Andres Legarra ◽  
Mary Kate Hollifield ◽  
Yutaka Masuda ◽  
Daniela Lourenco ◽  
...  
Keyword(s):  
Linear Regression ◽  
Single Step ◽  
Simulated Dataset ◽  
Linear Regression Method ◽  
Threshold Models ◽  
Genomic Information ◽  
Additive Variance ◽  
Breeding Values ◽  
Real Dataset ◽  
The Real

Abstract The objective of this study was to determine whether the linear regression (LR) method could be used to validate genomic threshold models used for analysis of categorical data. Statistics for the LR method were computed from estimated breeding values (EBVs) using the whole and truncated datasets. The method was tested using simulated and real chicken datasets. The simulated dataset included ten generations of 4,500 birds each; genotypes were available for the last three generations. Each animal was assigned a continuous trait, which was converted to a binary score assuming an incidence of failure of 7%. The real dataset included the survival status of 186,596 broilers (mortality rate equal to 7.2%) and genotypes of 18,047 birds. Both datasets were analyzed using Best Linear Unbiased Predictor (BLUP) or single-step GBLUP (ssGBLUP). The whole dataset included all phenotypes available, whereas in the partial dataset, phenotypes of the most recent generation were removed. In the simulated dataset, the accuracies based on the LR formulas were 0.45 for BLUP and 0.76 for ssGBLUP, whereas the correlations between true breeding values and EBVs were 0.37 and 0.65, respectively. The gain in accuracy by adding genomic information was overestimated by 0.09 when using the LR method compared to the true increase in accuracy. However, when the estimated ratio between the additive variance computed based on pedigree only and on pedigree and genomic information was considered, the difference between true and estimated gain was less than 0.02. Accuracies of BLUP and ssGBLUP with the real dataset were 0.41 and 0.47, respectively. Smaller improvements in accuracy when using ssGBLUP with the real dataset were due to population structure and lower heritability. The LR method is a useful tool for estimating improvements in accuracy of EBVs due to the inclusion of genomic information when traditional validation methods are not applicable.

Rainfall Prediction in South-Eastern Part of Bangladesh by Linear Regression Method

2018 ◽  
Vol 6 (6) ◽  
pp. 119
Author(s):  
Mohammad Shohidul Islam ◽  
Sultana Easmin Siddika ◽  
S M Injamamul Haque Masum
Keyword(s):  
Linear Regression ◽  
Correlation Coefficients ◽  
Rain Gauge ◽  
Regression Method ◽  
Linear Regression Method ◽  
Climate Data ◽  
Rainfall Forecasting ◽  
Statistical Parameters ◽  
Time Duration ◽  
Rainfall Prediction

Rainfall forecasting is very challenging task for the meteorologists. Over the last few decades, several models have been utilized, attempting the successful analysing and forecasting of rainfall. Recorded climate data can play an important role in this regard. Long-time duration of recorded data can be able to provide better advancement of rainfall forecasting. This paper presents the utilization of statistical techniques, particularly linear regression method for modelling the rainfall prediction over Bangladesh. The rainfall data for a period of 11 years was obtained from Bangladesh Meteorological department (BMD), Dhaka i.e. that was surface-based rain gauge rainfall which was acquired from 08 weather stations over Bangladesh for the years of 2001-2011. The monthly and yearly rainfall was determined. In order to assess the accuracy of it some statistical parameters such as average, meridian, correlation coefficients and standard deviation were determined for all stations. The model prediction of rainfall was compared with true rainfall which was collected from rain gauge of different stations and it was found that the model rainfall prediction has given good results.

An alternative model of the vibronic coupling in octahedral complexes

1988 ◽  
Vol 53 (6) ◽  
pp. 1134-1140
Author(s):  
Martin Breza ◽  
Peter Pelikán
Keyword(s):  
Linear Regression ◽  
Alternative Model ◽  
Potential Surface ◽  
Analytical Formula ◽  
Coupling Model ◽  
Regression Method ◽  
Vibronic Coupling ◽  
Linear Regression Method ◽  
Active Systems ◽  
Jahn Teller

It is suggested that for some transition metal hexahalo complexes, the Eg-(a1g + eg) vibronic coupling model is better suited than the classical T2g-(a1g + eg) model. For the former, alternative model, the potential constants in the analytical formula are evaluated from the numerical map of the adiabatic potential surface by using the linear regression method. The numerical values for 29 hexahalo complexes of the 1st row transition metals are obtained by the CNDO/2 method. Some interesting trends of parameters of such Jahn-Teller-active systems are disclosed.

The Accuracy Assessment of Weapon System Using Grey Linear Regression Method

2012 ◽  
Vol 268-270 ◽  
pp. 1809-1813
Author(s):  
Dai Yu Zhang ◽  
Bao Wei Song ◽  
Zhou Quan Zhu
Keyword(s):  
Linear Regression ◽  
Fuzzy Theory ◽  
Accuracy Assessment ◽  
Regression Method ◽  
Small Sample ◽  
Easy Access ◽  
Linear Regression Method ◽  
Grey Theory ◽  
Weapon System ◽  
Complex Engineering

The accuracy assessment of weapon system is always a complex engineering. How to make the most of the information given in only a few tests and obtain reasonable estimate is always a problem. Based on the fuzzy theory and grey theory, a grey linear regression method is presented. From the numerical example, we can see that this method provides an easy access to deal with data in small sample case and may have potential use in the analysis of weapon performance.

Perancangan Aplikasi Prediksi Jumlah Pendaftar Siswa Baru Dengan Metode Regresi Linier (Studi Kasus: SMA RK Deli Murni Bandar Baru)

2020 ◽  
Vol 3 (3) ◽  
pp. 330-334
Author(s):  
Novita Ria Lase ◽  
Fristi Riandari
Keyword(s):  
Linear Regression ◽  
Regression Method ◽  
Point Of View ◽  
Computational Techniques ◽  
Linear Regression Method ◽  
Simple Linear Regression ◽  
Tuition Fees ◽  
Use Of Data ◽  
New Students ◽  
New Student

The problem of the SMA RK Deli Murni Bandar Baru school is to predict how many facilities that need to be provided for new students such as chairs, tables and others. This study discusses the prediction of the number of new student registrants at SMA RK Deli Murni Bandar Baru based on the amount of tuition fees using a simple linear regression method. From a commercial point of view, the use of data mining can be used to handle the explosion of data volumes, using computational techniques can be used to produce information needed which is an asset that can increase the competitiveness of an institution. Prediction is almost the same as classification and estimation, except that in the prediction the value of the results will be in the future. This system can be used to predict the number of applicants in the following year to help the school. The advantage is that this simple linear regression method is very simple so that it is easy to calculate and use. Saves the time needed to solve problems, especially those that are very complex.

An application of QSRR approach and multiple linear regression method for lipophilicity assessment of flavonoids

2019 ◽  
Vol 164 ◽  
pp. 681-689 ◽  
Author(s):  
Mariusz Zapadka ◽  
Mateusz Kaczmarek ◽  
Bogumiła Kupcewicz ◽  
Przemysław Dekowski ◽  
Agata Walkowiak ◽  
...  
Keyword(s):  
Linear Regression ◽  
Multiple Linear Regression ◽  
Regression Method ◽  
Linear Regression Method ◽  
Multiple Linear Regression Method

A NEW PARAMETER ESTIMATION METHOD FOR ONLINE SOFT TISSUE CHARACTERIZATION

2016 ◽  
Vol 16 (08) ◽  
pp. 1640019 ◽  
Author(s):  
JAEHYUN SHIN ◽  
YONGMIN ZHONG ◽  
JULIAN SMITH ◽  
CHENGFAN GU
Keyword(s):  
Soft Tissue ◽  
Linear Regression ◽  
Tissue Characterization ◽  
Soft Tissues ◽  
Unscented Kalman Filter ◽  
Estimation Method ◽  
Regression Method ◽  
Linear Regression Method ◽  
Non Linear

Dynamic soft tissue characterization is of importance to robotic-assisted minimally invasive surgery. The traditional linear regression method is unsuited to handle the non-linear Hunt–Crossley (HC) model and its linearization process involves a linearization error. This paper presents a new non-linear estimation method for dynamic characterization of mechanical properties of soft tissues. In order to deal with non-linear and dynamic conditions involved in soft tissue characterization, this method improves the non-linearity and dynamics of the HC model by treating parameter [Formula: see text] as independent variable. Based on this, an unscented Kalman filter is developed for online estimation of soft tissue parameters. Simulations and comparison analysis demonstrate that the proposed method is able to estimate mechanical parameters for both homogeneous tissues and heterogeneous and multi-layer tissues, and the achieved performance is much better than that of the linear regression method.

scholarly journals Accident Prediction Model at Unsignalized Intersection Using Multiple Linear Regression Method

2017 ◽  
Vol 1 (2) ◽  
pp. 379-389
Author(s):  
N. K. Oghoyafedo ◽  
J. O. Ehiorobo ◽  
Ebuka Nwankwo
Keyword(s):  
Linear Regression ◽  
Prediction Model ◽  
Multiple Linear Regression ◽  
Regression Method ◽  
Linear Regression Method ◽  
Design Standards ◽  
Accident Prediction ◽  
Geometric Characteristics ◽  
Unsignalized Intersections ◽  
Multiple Linear Regression Method

The issue of road accidents is an increasing problem in developing countries. This could be due to increasing road traffic/vehicle occupancy, geometric characteristics and road way condition. The factors influencing accidents occurrence are to be analysed for remedies. The purpose of this research is to develop an accident prediction model as a measure for future study, aid planning phase preceding the designed intervention, enhance the production of updated design standards to enable practitioners design unsignalized intersection for optimal safety, reduce the number of accidents at unsignalized intersections. Five intersections were selected randomly within Benin City and traffic count carried out at these intersections as well as geometric characteristics and roadway conditions. The prediction model was developed using multiple linear regression method and the standard error of estimate was computed to show how close the observed value is to the regression line. The model was validated using coefficient of multiple determination. The establishment of the relationship between accidents and traffic flow site characteristics on the other hand would enable improvement to be more realistically accessed. This study will also enhance the production of updated design standards to enable practitioners design unsignalized intersection for optimal safety, reduce the number of accidents at unsignalized intersections.

Sign in / Sign up

Export Citation Format

Share Document