CHAPTER-V GENOTYPE-ENVIRONMENT INTERACTION: ANALYSIS FOR NEIGHBOUR EFFECTS
|
|
- Joseph Newton
- 5 years ago
- Views:
Transcription
1 CHAPTER-V GENOTYPE-ENVIRONMENT INTERACTION: ANALYSIS FOR NEIGHBOUR EFFECTS 5.1 Introduction Designs of experiments deal with planning, conducting, analyzing and interpreting tests to evaluate the factors that control the value of a parameter or group of parameters. As first step of experiments is planning, construction of design in statistical terminology. Hence, previous two chapters are devoted to the construction of neighbour design using different methods of construction. Then one shall be interested to analyze the data for the significance. Analysis of data shall be discussed in this chapter to obtain stability parameters for neighbour designs. Here is considered the earlier part i.e. analysis of the design, later one shall be discussed in the next chapter. Although plant breeders were aware of the differential responses of improved genotypes to changes in environments long back but statistical analyses were developed in early sixties. A dynamic approach to interpretation of varying environments was developed by Finlay and Wilkinson (1963). It leads to the discovery that the components of a genotype and environment interaction were linearly related to environmental effects, when these effects were measured on the same scale as the genotypic effects. The regression technique of Finlay and Wilkinson was improved upon by Eberhart and Russell (1966) by adding another stability parameter, viz. the deviation from regression and provided a fresh approach to GE interaction analysis. The approaches given by Finlay and Wilkinson (1963) and Eberhart and Russell (1966) are purely statistical and the components of this analysis have not been related to parameters in a biometrical genetical model. The second approach, based on fitting of models which specify contributions of genetic, environmental and genotype-environment interactions to generation means, variances, etc., was developed by Bucio Alanis (1966). 106
2 Bucio Alanis and Hill (1966) extended the above model by including some parental effect for averaged dominance over all environments. Perkins and Jinks (1968a) extended the technique to include many inbred lines and considered analysis of GE interaction with different angle. The approach has come to be known as joint regression analysis as each genotype is regressed onto the environments, which is measured by the joint effect of all genotypes. This technique has been widely used to measure the contribution of genotypes to the GE interactions and predicting performance of genotypes over environments and generations. Some major weakness of the statistical theory of regression approach developed by these workers were pointed out by raising the following basic issues, (i) the fundamental statistical assumptions were not usually satisfied and that the choice of the sums of squares and degrees of freedom from which the regression components were subtracted was not proper, (ii) the use of mean yield under each environment as the environmental value of the particular set of genotypes as an independent variable which was, in fact, not independent of the phenotypic variable regressed onto it. With these considerations, Freeman and Perkins (1971) developed an approach, known as modified approach. Digby (1979) developed a modified regression approach wherein he improved upon the adjustment of Patterson by introducing a sensitivity parameter. Finney(1980), Patterson and Silver(1980) & Patterson(1980) further extended the approach for such cases. Laxmi (1992) found a relationship between different regression models and different components in a model by a graphic method and had shown how these are interlinked with each other. She also developed a new line of stability analysis using indicator variable so as to judge adaptability of varieties to meet potential yield when some observations are missing. Finlay and Wilkinson (1963) worked on classical block model where assumptions are that the response on a plot to a particular treatment does not affect the response on the neighbouring plots and the fertility associated with plots in a block is constant. However, in many fields of agricultural research, the treatment applied to one experimental plot in a block may affect the response on the neighbouring plots if the blocks are linear with no 107
3 guard areas between the plots. The estimates of treatment differences may therefore deviate because of interference or competition from neighbouring units. So after 20 years, Wilkinson et al. (1983) proposed a method called nearest neighbour (NN) analysis in which local trends are removed by taking second differences or partial differences. B. L. Misra, Bhagwandas and Nutan (1991) gave several methods of construction of generalized neighbour design of equal and unequal block sizes and an outline of analysis of generalized neighbour design has been suggested. Azais et al. (1993) defined the block design for competition effects to be balanced and he also gave an outline of analysis for models with no neighbour effect, one-sided neighbour effect and two-sided neighbour effects. Bailey (2003) considered analysis of only one-side neighbour effects. Jaggi Gupta and Ashraf (2006) suggested analysis of complete block designs partially balanced for neighbouring competition effects and have also been obtained efficiency of neighbour effects. Pateria et al.(2007) considered a series of block designs by putting N-1 MOLS with N treatments one below another and had obtained designs for N5 & N7 with necessary analytical methods. Pateria, Jaggi and Varghese(2009) suggested analysis for self-neighboured strongly balanced block designs and have given analysis for the designs which are variance balanced for estimating the contrast pertaining to direct effects and also variance balanced for estimating the contrast pertaining to neighbour (left and right) effects. Though Wilkinson et al. (1983) proposed a method called nearest neighbour (NN) analysis in which local trends are removed by taking second differences or partial differences but they do not consider the competition effects or interference effects for finding out the stability measures of variety. Here emphasis is given on finding out the stability measures for competition i.e. neighbour designs. In this chapter analysis is considered for neighbour design for GE interaction. 108
4 5.2 Eberhart and Russell (1966) Model Before discussing the model given by Eberhart and Russell (1966) let us present the approach to GE interaction analysis given by Finlay and Wilkinson (1963). In this technique, the mean yield of all genotypes for each location is considered for quantitative grading of the environments and the linear regression of the mean values for each genotype onto the mean values for environments is estimated. The model is defined as y ij μ i + B i I j + δ ij + e ij (5.2.1) where y ij is the ith genotype mean in the jth environment, i1,2,...,v; j1,2,...,b; μ i is the overall mean of the ith genotype; B i is the regression coefficient that measures the response of the genotype of varying environments, δ ij is the deviation from regression of the ith genotype at the jth environment; e ij is the error such that E(e ij ) 0 and Var (e ij ) V and I j is the environmental index defined by I j - (5.2.2) such that 0 Hence, the regression coefficients, B i may be estimated as B i / (5.2.3) and it is considered as the first stability statistic. (The analysis of variance (ANOVA) of the model has been presented in table 5.2.1) 109
5 Table Analysis of Variance (Finlay and Wilkinson, 1963 Model) Source of Degrees of Sum of Square Mean Sum of Variation Freedom Square Replication with b(n-1) Environments pooled _ environments over Genotypes(G) (v-1) CF M G Environments (E) (b-1) ( ) 2 / b M E GE interaction (GE) (v-1)(b-1) - M GE Regressions (v-1) / M R Deviation from regression (v-1)(b-2) GE SS Regression SS M d Error b(v-1)(n-1) Pooled M e On the basis of the empirical findings Eberhart and Russell (1966) developed their regression model on similar lines that there exists a linear relationship between phenotype 110
6 and environment and measured its effect on the character under study. However, they adopted another approach to obtain the phenotypic regression (b i ) of the value y ij on the environment E j, as against to genotypic regression (B i ) of g ij on E j as formulated by Finlay and Wilkinson. The model considered by Eberhart and Russell may be written as y ijk μ i + b i E j + δ ij + e ijk (5.2.4) where y ijk is the phenotypic value of the ith genotype at the jth environment in the kth replicate (i 1,2,...,v; j 1,2,...,b; k 1,2,...,n), μ i is the mean of the ith genotype over all the environments, b i is the regression coefficient that measures the response of ith genotype to the varying environments, E j is the environmental index obtained as - such that 0, δ ij is the deviation from regression of the ith genotype in the jth environment, and is the random component. The ANOVA for this model is shown in table With this model, the sums of the squares due to environments and GE interaction are partitioned into environments (linear), GE interaction (linear) and deviations from the regression model. Table Analysis of Variance (Eberhart and Russell, 1966 Model) Source of Degrees of Sum of Square Mean Sum Variation Freedom of Square Genotypes(G) (v-1) CF M G Environments (E) (b-1) CF M E GE interaction (GE) (v-1)(b-1) CF M GE 111
7 Environment (linear) (El) 1 / M El GE (linear) (GEl) (v-1) / - SS Env. (linear) M GEl Pooled Deviation (d) v(b-2) M d Deviation due to genotype i (v-2) - ( / M g Pooled Error b(v-1)(n-1) Pooled over environments M e With this approach, the first stability parameter is a regression coefficient, b i which can be estimated by b i / By using the estimates of the parameters, performance of each genotype can be predicted as ij + b i E j, where is an estimate of the µ i. The deviation δ ij [ (y ij - i)]can be squared and summed to provide an estimate of another stability parameter ( ) as [ / (b-2)] / n (5.2.5) where / n is the estimate of the pooled error, and [ ij - ] - ( / (5.2.6) two parts: The model provides a means of partitioning the GE interaction of each genotype into 112
8 (i) (ii) the variation due to the response of genotype to varying environmental indices (sums of square due to regression), and the unexplainable deviation from regression on the environmental indices. Further, they define that the stable variety will be one with b i 1.0 and 0; and the null hypothesis H 0 : µ 1 µ 2 µ m can be tested by the F-test (approximately) F M G / M d with homogeneous deviation mean squares, being M d the pooled deviations. The hypothesis that there are no genetic differences among phenotypes for their regression on the environmental index H 0 : β 1 β 2 β m can be tested by the F-test F M EI / M d The deviations from regression for each genotype can be further tested by F [ / ( b-2] / Pooled error. Thus, in this approach one can see that two measures of sensitivity of the genotype to changes on environment are worked out: (i) the linear sensitivity measure in terms of the linear regression coefficient, b i, of the ith genotype to the environmental change, and (ii) the non linear sensitivity measure in terms of the deviation from regression mean square, i.e. / (b-2). Hence, the second stability parameter ( played a very important roleas the estimated variance,, is a function of the number of environments, so several environments with minimum replications per environment are necessary to obtain reliable estimates of. The pooled deviations were partitioned genotype wise accordingly so to examine their stability. 113
9 5.3 Perkins and Jinks (1968) Model Perkins and Jinks (1968) extended the technique of Bucio Alanis (1966) and Bucio Alanis and Hill (1966) by improving their models. They describe the performance of the ith genotype in the jth environment as given by the model y ij μ + d i + E j + g ij + e ij (5.3.1) where μ is the general mean over all genotypes and environments; d i is the additive effect of the ith genotype ; E j is the effect of the jth environment; g ij is the GE interaction of the ith genotype with jth environment; and e ij is the error terms. Since d i, E j and g ij are fixed effects over the jth environment and therefore, 0; 0 and 0 (5.3.2) If the ith genotype is regressed onto the jth environment, one can write g ij B i E j + δ ij (5.3.3) where B i is the linear regression coefficient for the ith genotype and δ ij is the deviation from regression for the ith genotype in the jth environment. Hence the model in (5.3.1) can be written as y ij μ + d i + (1 + B i ) E j + δ ij + e ij (5.3.4) By following the approach, two aspects of phenotypes are recognized: (i) linear sensitivity to change in environment as measured by the regression coefficient, B i and (ii) non linear sensitivity to environmental changes which is expressed by δ ij. The ANOVA for the model has been presented in table
10 Table Analysis of Variance (Perkins and Jinks, 1968 Model) Source of Degrees of Sum of Square Mean Variation Freedom Sum of Square Genotypes(G) (v-1) b M G Environments (E) (b-1) v M E GE interaction (GE) (v-1)(b-1) - SSG - SSE M GE Heterogeneity M H between regression (v-1) (H) Reminder (v-1)(b-2) M d (d) Error b(v-1)(n-1) Pooled M e It may be seen from the ANOVA table that the analysis of variance consists of two parts: 1. A conventional analysis of variance so as to check whether GE interaction is significant, and 115
11 2. To test whether GE interaction is a linear function of the additive environmental component. For this purpose the GE interaction is partitioned further into two parts: (i) (ii) heterogeneity between regression sum of squares, and the reminder sum of squares. The two components of GE interaction can be tested for their significance against M e, which is the error mean square due to within genotype, and within environmental variations averaged over all environments. If either of the heterogeneity regression mean square, the remainder mean square or both are significant, one may conclude that GE interactions are significant. If heterogeneity mean square alone is significant one may derive the finding that GE interactions for each genotype may be treated as linear regression (for genotype) on the environmental values. If the remainder mean square alone is significant, it may be assumed that there is no evidence of any relationship between the GE interactions and the environmental values and no prediction can be made with this approach. When both the components are significant the usefulness of any prediction will depend on the relative magnitude of these mean squares. However, in case of significance of heterogeneity mean square compared with the remainder mean square, the GE interaction may be useful for prediction and shows its considerable importance in such cases. In this approach, two measures of sensitivity of the genotype to changes on environment are calculated: (i) the linear regression coefficient, B i, of the ith genotype to the environmental measure giving as a measure of the linear sensitivity, and (ii) the deviation from regression mean square, / (b-2) a measure of the nonlinear sensitivity. 116
12 If model of Eberhart and Russell (1966) and model of Jinks and Perkins (1968) are compared then the following relationships hold between the parameters: μ i (μ + d i ); b i (1 + B i ) and δ ij. δ ij (5.3.5) The relationship between two regression coefficients, B i and b i under these models occurs because in model the GE component, g ij, is regressed onto E j values and the regression of E j on itself has got already a slope of unity, while in model only the GE component, g ij, is regressed onto E j. The estimates of the various parameters can be obtained as (5.3.6) where is the mean yield of all the genotypes in all environments; d i can be obtained as the deviation from the mean yield for the ith genotype, i.e. - (5.3.7) and E j environmental index can be obtained as the deviation from the mean yield for the jth environment, i.e. E j - (5.3.8) 5.4 Freeman and Perkins (1971) Model Some statistical shortcomings have been brought out against the regression models given by Eberhart and Russell (1966) and Jinks and Perkins (1968a) as described earlier. One may discuss the major two as under: 1. The environment (linear) sum of squares which has the same sum of squares for environments with (b-1) degrees of freedom in Jinks and Perkins (1968a) model has been allocated only one degree of freedom for the same sum of squares in Eberhart and Russell (1966) model, which amounts to an over-estimation of the 117
13 corresponding mean square. According to the statistical rules, any sum of squares has a unique number of degrees of freedom and these cannot vary if a different partitioning scheme is adopted. 2. The use of non-independent measure of environment, that is, the mean of all varieties in that environment, leads to a statistical invalidity of the model. With these observations, Freeman and Perkins (1971) developed a new approach having rectification to the shortcomings. The first short-coming which relates to analysis part was removed by a modification in the method of analysis. The second relates to the model and it was met by introducing an independent measure of the environments. Freeman and Perkins (1971) considered the model as y ijk μ + d i + E j + g ij + e ijk (5.4.1) where y ijk stands for the performance of the ith genotype in the jth environment and in the kth replicate, and the remaining stands as specified in (5.3.1). Using the convention that y ij. refers to the total over the suffix replaced by a dot, whereas refers to the mean over the suffix omitted; more than one suffix can be replaced or omitted. Further, μ is estimated by, d i by (..), E j by (.. ) and g ij by ( +.). The ANOVA has been presented in Table The sums of squares can also be replaced by appropriate summations of their respective estimates, i.e. the sum of squares being expressible either as - or as
14 Table Analysis of Variance (Freeman and Perkins, 1971 Model) Source of Variation Degrees of Freedom Sum of Square Genotypes(G) (v-1) - Environments (E) (b-1) - Combined regression 1 - Residual b-2 Env. SS Reg. SS. GE interaction (GE) (v-1)(b-1) - + Heterogeneity of regression (v-1) - / Reminder (v-1)(b-2) GE SS Het. Reg. SS Error vb(n-1) - / n 119
15 The total variation within genotypes is obtained by summing up the E and GE components. Partitioning GE interaction into m groups is not possible as (v-1)(b-1) df are not divisible by v. As all the components are orthogonal, partitioning of GE sum of squares into (v-1) components attributable to various genotypes has been considered by following up the approach differently, which is discussed as under: Let x j be the measure of the environment and it is the same for all the genotypes. In the foregoing works it was taken x j to be or ( ), but it is not valid for x j to be linear function of because the linear regression of on x j is considered. This may be expressed as: B i (x j ) + error term (5.4.2) where is the mean of x j. Assuming (x j ) as Z j, one can estimate B i by b i, where b i n C xy / n C xx (say). After some adjustments one may have the sum of squares for regression, with one df, by n /. Subtracting this expression from gives the deviation from regression within the ith genotype as a sum of squares with (b-2) df. The combined regression line over all genotypes has slope / r m / r m C xx (say) (5.4.3) The overall regression sum of squares is vn 2 C xx, i.e. v n /, the square of a linear function of, and thus part of the environment sum of squares. 5.5 Laxmi (1992) Model From discussions of the methods given by Eberhart and Russell (1966), Perkins and Jinks (1963), Freemen and Perkins (1971) and several other workers, it is quite apparent that phenotypic expression, y ij of ith genotype in the jth environment depends on three genotypic properties; a mean expression, a linear response to environment and residual 120
16 deviation from regression. Here one may remember that the estimates of various parameters are given as, d i., E j. and +. (5.5.1) with the assumption that E (e ij ) 0 and Var (e ij ) constant. Generally, there is observed heterogeneity of variance due to environments, which may arise due to their differences in nature. Let us assume that Var (e ij ) V j for jth environment and hence, Laxmi(1992) gave the model by B i (x j ) + e ij B i Z j + e ij (5.5.2) where Z j (x j ), x j is a measure on the environment j, is the mean of x j, B i the regression coefficient and e ij the error term. Now we may consider a set of positive numbers, W j, to be used as weighting factor for a weighted linear regression, and let W j V -1 j.without loss of generality, we may assume that the Z j have been so chosen that 0 (5.5.3) After considering residuals and applying the method of least squares, we have / and / (5.5.4) Relations remain valid even when the W j are an arbitrary set of positive numbers, rather than the reciprocals of variances, V j and will always provide unbiased estimates corresponding to population parameter. Anscombe and Turkey (1963) have examined such solutions and pointed out that these estimates would not enjoy the maximum variance 121
17 property of the least squares analysis. The same holds true when arbitrary W j are taken into consideration. However, the situation is entirely analogous to the estimation of components of variance from an analysis of variance. The weighted analysis of variance for the model is shown in Table If the error variances are very different from each other then all the analyses made by regression approach may fail, with this approach it is possible to bring out a solution to the problem and can go for further analyses of GE interaction and stability. Table Analysis of Variance (Laxmi, 1992 Model) Source of Degrees of Sum of Square Mean Sum Variation Freedom of Square Genotypes(G) (v-1) M G Environments (E) (b-1) M E GE interaction (GE) Linear Regression (v-1)(b-1) (TSS SSG SSE) SSGE M GE (v-1) SSLR M LR (LR) Residual (v-1)(b-2) (SSGE SSLR) M r (r) Error b(v-1)(n-1) Pooled M e 122
18 If a genotype is included for which B i 0, i.e. it is completely insensitive to the environmental effect then the data for the genotype should not be used to estimate environmental effects in ordinary analysis. Such data are not to be excluded from the present analysis and one should proceed as usual with the modified methods. These were the models discussed by authors for GE analysis. To include the competition effect one has to consider models with neighbour effects. 5.6 Proposed Block Model for GE Interaction with Two-Sided Neighbour Effects In an experiment on competition studies, the effect of a genotype/treatment applied to a plot can be written as the total sum of direct effect due to the treatment applied to a given plot, left neighbour effect due to the competition with the treatment applied to the immediate left neighbour plot and right neighbour effect due to the competition with the treatment applied to the immediate right neighbour plot. Therefore, the competition treatments constitute an ordered triplet. Blocks of the competition designs are the sets of these triplets arranged in sequence. Several authors had worked on neighbouring competition effects for the analysis of genotype/treatment effects but no reference till date is available on analysis of GE interaction considering neighbour effects. The model with differentiated two- sided neighbour effects for GE interaction is: Y ijl μ + τ (i, j) + δ (i - 1, j) + ρ (i + 1, j) + E j + g ij + e i jl (5.6.1) where Y ijl is the response from the ith plot in the jth block ( i 1,2,...,k; j 1,2,..., b; l 1,2,,n), μ is the general mean effect, τ (i, j) is the direct effect of the genotype/treatment in the ith plot of jth block, E j is the effect of jth environment/block, δ (i - 1, j) is the left neighbour effect due to the genotype/treatment in the (i 1)th plot of jth block, ρ (i+1,j) is the right neighbour effect due to the treatment/ genotype in the (i + 1)th plot of jth block, g ij is the GE interaction of the ith genotype with jth environment and e ijl are error terms independently and normally distributed with mean zero and variance σ 2. Since τ i,j, δ (i 1),j, ρ (i + 1),j, E j and g ij are fixed effects and therefore, 123
19 0; 0 and 0 (5.6.2) If the ith genotype is regressed onto the jth environment, one can write g ij B i E j + η ij (5.6.3) where B i is the linear regression coefficient for the ith genotype and η ij is the deviation from regression for the ith genotype in the jth environment. Considering this set-up, the joint information matrix for model (5.6.1) for estimating different effects by generalized least squares is obtained as follows: C where M 1 is the v x v incidence matrix pertaining to direct versus left treatments, M 2 is the v x v incidence matrix pertaining to direct versus right treatments, M 3 is the v x v incidence matrix pertaining to left versus right treatments, N 1 is the v x b incidence matrix pertaining to direct treatments versus blocks, N 2 is the v x b incidence matrix pertaining to left treatments versus blocks, N 3 is the v x b incidence matrix pertaining to right treatments versus blocks. Further R τ diagonal (r 1, r 2,, r v ); r i (i 1, 2, v) being the number of times the i-th treatment appears in the design, 124
20 R δ diagonal (r 11, r 12,...,r 1v ) being the number of times the treatments in the design has i-th treatment as left neighbour, R ρ diagonal (r 21, r 22, r 2v ) being the number of times the treatments in the design has i-th treatment as right neighbour. K diagonal (k 1, k 2, k b ) The above 3v x 3v information matrix (C) for estimating the different effects in a block design setting are non-negative definite with row and column sums equal to zero. For the multi-environment experiments which are conducted in incomplete design for v genotypes in b environments with border plots as discussed in chapter III and IV with parameters vbs 2 +s+1, rks+1, λ1, we obtained R τ R δ R ρ ri; and [(r-1) I+J] where I is an identity matrix of order v x v and J is a matrix of order v x v with all elements unity. Further, for neighbour design of OS2 series M 1 J rj, J 2 vj, M 1 I M 1, J rj, I,, and r k. Therefore, the joint information matrix for estimating different effects for neighbour designs of OS2 series, reduces to C 1 (5.6.4) 125
21 using the result given by Dey(1986, p.93) which shows that for a BIBD R NK -1 N [I v -1 J ] (5.6.5) Then matrix C 1 becomes C 1 (5.6.6) Again this 3v x 3v matrix i.e. C 1 is non-negative definite with zero row and column sums. The information matrix for estimating the direct effects of treatments i.e. C τ can be obtained as given below: C τ A 1 B 1 D 1 (5.6.7) where A 1 [I v -1 J ], (5.6.8) B 1 (5.6.9) D 1 (5.6.10) As D 1 is a singular matrix, therefore its g-inverse has been found out. So D 1 126
22 (5.6.11) Now, consider B 1 D 1 (5.6.12) (5.6.13) (5.6.14) (5.6.15) ( because M 1 J rj, M 1 I M 1, and r k) On post multiplying by, we get B 1 D 1 x (5.6.16) (5.6.17) (5.6.18) (5.6.19) On simplification, we obtain B 1 D 1 (5.6.20) Finally, C τ so obtained is 127
23 C τ A 1 B 1 D 1 [I v -1 J ] - (5.6.21) I - (5.6.22) Thus C τ τ 1 Q 1 (5.6.23) are the reduced normal equations for estimating the direct treatment effects and solutions of these equations provide, where (,,, )' and Q 1 T 1 N 1 K -1 B (5.6.24) further T 1 is the vector of total of direct treatment effects and N 1 is the v x b incidence matrix pertaining to direct treatments versus blocks. Now consider the joint information matrix, for estimating different effects for neighbour designs of OS2 series, C 2, which is obtained by interchanging first & second rows and then interchanging first & second columns of C 1 matrix. Therefore, C 2 obtained is C 2 (5.6.25) using the result given by Dey(1986), the matrix C 2 becomes C 2 (5.6.26) 128
24 Again this 3v x 3v matrix i.e. C 2 is non-negative definite with zero row and column sums. The information matrix for estimating the left treatment effects i.e. C δ can be obtained as given below: C δ A 2 B 2 D 2 (5.6.27) where A 2, (5.6.28) B 2 (5.6.29) D 2 (5.6.30) As D 2 is a singular matrix, therefore, its g-inverse has been found out. So D 2 (5.6.31) (5.6.32) Now, consider B 2 D 2 (5.6.33) (5.6.34) (5.6.35) (5.6.36) 129
25 ( because J rj, vj, I, and r k) On post multiplying by, we get B 2 D 2 x (5.6.37) (5.6.38) (5.6.39) (5.6.40) On simplification, we obtain B 2 D 2 (5.6.41) Finally, C δ so obtained is C δ A 2 B 2 D 2 [I v -1 J ] - (5.6.42) I - (5.6.43) Then C δ τ 2 Q 2 are the reduced normal equations for estimating the left treatment effects and solutions of these equations provide, i.e. vector of estimates of left treatment effects and Q 2 T 2 N 2 K -1 B where T 2 is the vector of total of left neighbours and N 2 is the v x b incidence matrix pertaining to left treatments versus blocks 130
26 Now again consider the joint information matrix C 1 for estimating different effects for neighbour designs of OS2 series C 1 (5.6.44) which is partitioned in such a way that A 3, (5.6.45) B 3 (5.6.46) D 3 (5.6.47) The information matrix for estimating the right effects of treatments i.e. C ρ can be obtained as given below: C ρ A 3 B 3 D 3 (5.6.48) As D 3 is a singular matrix, therefore, its g-inverse has been found out. So D 3 (5.6.49) (5.6.50) Now, consider B 3 D 3 (5.6.51) 131
27 (5.6.52) (because ) (5.6.53) (5.6.54) (5.6.55) On post multiplying by we get B 3 D 3 x (because ) Therefore, B 3 D 3 x (5.6.56) (5.6.57) (5.6.58) (5.6.59) On simplification, we obtain B 3 D 3 (5.6.60) Finally, C ρ so obtained is C ρ A 3 B 3 D 3 132
28 [I v -1 J ] - (5.6.61) I - (5.6.62) Then C ρ τ 3 Q 3 are the reduced normal equations for estimating the right treatment effects and solutions of these equations provide, i.e. vector of estimates of right treatment effects and Q 3 T 3 N 3 K -1 B where T 3 is the vector of total of right neighbours and N 3 is the v x b incidence matrix pertaining to right treatments versus blocks Putting back these estimates of direct treatment effects, left neighbour effects and right neighbour effects, separated above, in the model, the total sum of squares is partitioned into different components of variation constituting analysis of variance. Then ANOVA for GE Interaction with neighbouring effect is shown in Table With this model, the sum of squares due to environments and GE interaction are partitioned into environments (linear), GE interaction (linear) and deviations from the regression model. Table Analysis of Variance (GE interaction with Neighbouring Effect, Model) Source of Degrees of Sum of Square Mean Sum of Variation Freedom Square Environments (E) s 2 +s CF M E Genotypes(G) s M G Left Effect(L) s M L 133
29 Right Effect (R) s M R GE interaction (GE) s( s 2 +s ) CF M GE Environment (linear) (El) GE (linear) (GEl) 1 / s / - SS Env. (linear) M El M GEl Pooled Deviation (d) s(s 2 +s-1) M d Deviation due to genotype i s 2 +s-1 ( - M g Pooled Error s(s 2 +s+1)n Pooled over environments * n is the number of times the whole BIBD is repeated. Here is the adjusted treatment sum of square for direct treatments is the adjusted treatment sum of square for left neighbours and is the adjusted treatment sum of square for right neighbours For calculation purpose, these equations further can be replaced by sum of squares with summations (5.6.63) 134
30 (5.6.64) (5.6.65) If model proposed here is compared with the model given by Eberhart and Russell (1966), then the following relationship holds μ i μ + τ (i, j) + δ (i - 1, j) + ρ (i + 1, j) ; b i (1 + B i ) and δ ij. η ij (5.6.66) It may be seen from the ANOVA table that sum of squares of direct treatment effects is different from the one obtained by Eberhart and Russell as the earlier consists of the effects of neighbour treatments. It further results in differentiating sum of squares due to GE interaction considering neighbour effects in the proposed model, from that of considering only direct treatment effects, discussed by Eberhart and Russell (1966). Therefore, regression of genotype on environments and deviation due to genotype may be highly significant when tested against which is the pooled mean square error. This gives an idea for measure of stability parameters considering neighbour effects, which is going to be discussed in the following chapter. References Anscombe, F.J. and Tukey, J.W. (1963): The examination and analysis of residuals.,technometrics, 5, Azais, J.M., Bailey, R.A. & Monod, H.(1993): A catalogue of efficient neighbour designs with border plots. Biometrics, 49, Bailey, R.A. (2003): Designs for one- sided neighbour effects. Journal of the Indian Society Agriculture Statistics,56(3), Bucio Alanis, L. (1966). Environmental and genotype-environment components of variability I-Inbred Lines. Heredity, 31,
31 Bucio Alanis, L. and Hill, J. (1966).Environmental and genotype environmental components of variability II.Heterozygotes. Heredity, 21, Dey, A. (1986): Theory of Block Designs. Wiley Eastern Limited, New Delhi. Digby, P.G.N. (1979). Modified joint regression analysis for incomplete variety x environment data. J. Agric. Sci. Camb., 93, Eberhart, S.A. and Russel, W.A. (1966).Stability parameters for comparing varieties. Crop Sci., 6, Finlay, K.W. and Wilkinson, G.N. (1963).The analysis of adaptation in a plant breeding programme. Aust. J. Agri. Res., 14, Finney, D.G. (1980). The estimation of parameters by least squares by unbalanced experiments. J. Agric. Sci., Camb., 95, Freeman, G.H. and Perkins, J.M. (1971): Environmental and genotype-environmental components of variability VIII. Relation between genotype grown in different environments and measurement of these environments. Heredity, 27, Jaggi, Seema, Gupta, V.K., and Ashraf, Jawaid (2006): On block designs partially balanced for neighbouring competition effects. Journal of the Indian Statistical Association. Vol. 44, 1, Laxmi, R.R.,(1992): Genotype-Environment interaction: Its role in stability of crop varieties. Unpublished thesis, submitted to IASRI, New Delhi. Misra, B.L., Bhagwandas & Nutan (1991): Families of neighbour designs and their analysis. Communication Statistics Simulation, 20(2&3), Pateria, Dinesh Kumar, Jaggi, Seema, Varghese, Cini and Das, M.N.(2007): Incomplete non- circular block designs for competition effects. Journal of Statistics & Applications. Vol. 5 nos. 1&2, (New Series),
32 Pateria, Dinesh Kumar, Jaggi, Seema and Varghese, Cini (2009): Self-Neighboured Strongly Balanced Block Designs. Journal of the Indian Statistical Assoc. 47(1), Patterson. H.D. (1978). Routine least squares estimation of variety means in incomplete tables. J. Nat. Instt. Agric. Bot., 14, Patterson. H.D. (1980): Yield sensitivity and straw shortness in varieties of winter wheat.j. Nat. Instt. Agric. Bot., 15, Patterson, H. D. and Silvey, V. (1980). Statutory and recommend list trials of crop varieties in U.K. J. Roy. Stat. Soc., A.,143, Perkins, J. M. and Jinks, J. L. (1968). Environmental and genotype environmental components of variability, III. Multiple lines and crosses. Heredity, 23, Wilkinson, G. N., Eckert, S. R., Hancock, T. W. & Mayo, O. (1983).Nearest neighbour (NN) analysis of field experiments (with discussion). J. R. Statist. Soc. B 45,
Efficiency of NNBD and NNBIBD using autoregressive model
2018; 3(3): 133-138 ISSN: 2456-1452 Maths 2018; 3(3): 133-138 2018 Stats & Maths www.mathsjournal.com Received: 17-03-2018 Accepted: 18-04-2018 S Saalini Department of Statistics, Loyola College, Chennai,
More informationSAMPLING IN FIELD EXPERIMENTS
SAMPLING IN FIELD EXPERIMENTS Rajender Parsad I.A.S.R.I., Library Avenue, New Delhi-0 0 rajender@iasri.res.in In field experiments, the plot size for experimentation is selected for achieving a prescribed
More informationMUTUALLY ORTHOGONAL LATIN SQUARES AND THEIR USES
MUTUALLY ORTHOGONAL LATIN SQUARES AND THEIR USES LOKESH DWIVEDI M.Sc. (Agricultural Statistics), Roll No. 449 I.A.S.R.I., Library Avenue, New Delhi 0 02 Chairperson: Dr. Cini Varghese Abstract: A Latin
More informationNeighbor Balanced Block Designs for Two Factors
Journal of Modern Applied Statistical Methods Volume 9 Issue Article --00 Neighbor Balanced Block Designs for Two Factors Seema Jaggi Indian Agricultural Statistics Research Institute, New Delhi, India,
More informationNESTED BLOCK DESIGNS
NESTED BLOCK DESIGNS Rajender Parsad I.A.S.R.I., Library Avenue, New Delhi 0 0 rajender@iasri.res.in. Introduction Heterogeneity in the experimental material is the most important problem to be reckoned
More informationCOVARIANCE ANALYSIS. Rajender Parsad and V.K. Gupta I.A.S.R.I., Library Avenue, New Delhi
COVARIANCE ANALYSIS Rajender Parsad and V.K. Gupta I.A.S.R.I., Library Avenue, New Delhi - 110 012 1. Introduction It is well known that in designed experiments the ability to detect existing differences
More informationc. M. Hernandez, J. Crossa, A. castillo
THE AREA UNDER THE FUNCTION: AN INDEX FOR SELECTING DESIRABLE GENOTYPES 8 9 c. M. Hernandez, J. Crossa, A. castillo 0 8 9 0 Universidad de Colima, Mexico. International Maize and Wheat Improvement Center
More informationGENERALIZATION OF NEAREST NEIGHBOUR TREATMENTS
International Journal of Advances in Engineering & Technology, Jan 04 GENERALIZATION OF NEAREST NEIGHBOUR TREATMENTS USING EUCLIDEAN GEOMETRY Rani S, Laxmi R R Assistant Professor, Department of Statistics
More informationMULTIVARIATE ANALYSIS OF VARIANCE
MULTIVARIATE ANALYSIS OF VARIANCE RAJENDER PARSAD AND L.M. BHAR Indian Agricultural Statistics Research Institute Library Avenue, New Delhi - 0 0 lmb@iasri.res.in. Introduction In many agricultural experiments,
More informationSOME ASPECTS OF STABILITY OF CROP VARIETIES
SOME ASPECTS OF STABILITY OF CROP VARIETIES V.K. Bhatia I.A.S.R.I., Library Avenue, New Delhi 110 01 vkbhatia@iasri.res.in 1. Introduction The performance of a particular variety is the result of its genetic
More informationDesigns for asymmetrical factorial experiment through confounded symmetricals
Statistics and Applications Volume 9, Nos. 1&2, 2011 (New Series), pp. 71-81 Designs for asymmetrical factorial experiment through confounded symmetricals P.R. Sreenath (Retired) Indian Agricultural Statistics
More informationAnalysis of Variance and Co-variance. By Manza Ramesh
Analysis of Variance and Co-variance By Manza Ramesh Contents Analysis of Variance (ANOVA) What is ANOVA? The Basic Principle of ANOVA ANOVA Technique Setting up Analysis of Variance Table Short-cut Method
More informationBALANCED INCOMPLETE BLOCK DESIGNS
BALANCED INCOMPLETE BLOCK DESIGNS V.K. Sharma I.A.S.R.I., Library Avenue, New Delhi -110012. 1. Introduction In Incomplete block designs, as their name implies, the block size is less than the number of
More informationBalanced Treatment-Control Row-Column Designs
International Journal of Theoretical & Applied Sciences, 5(2): 64-68(2013) ISSN No. (Print): 0975-1718 ISSN No. (Online): 2249-3247 Balanced Treatment-Control Row-Column Designs Kallol Sarkar, Cini Varghese,
More informationAn Incomplete Block Change-Over Design Balanced for First and Second-Order Residual Effect
ISSN 66-0379 03, Vol., No. An Incomplete Block Change-Over Design Balanced for First and Second-Order Residual Effect Kanchan Chowdhury Professor, Department of Statistics, Jahangirnagar University Savar,
More informationLecture 08: Standard methods. Bruce Walsh lecture notes Tucson Winter Institute 9-11 Jan 2013
Lecture 08: G x E: Genotype-byenvironment interactions: Standard methods Bruce Walsh lecture notes Tucson Winter Institute 9-11 Jan 2013 1 G x E Introduction to G x E Basics of G x E G x E is a correlated
More informationROW-COLUMN DESIGNS. Seema Jaggi I.A.S.R.I., Library Avenue, New Delhi
ROW-COLUMN DESIGNS Seema Jaggi I.A.S.R.I., Library Avenue, New Delhi-110 012 seema@iasri.res.in 1. Introduction Block designs are used when the heterogeneity present in the experimental material is in
More informationGENOTYPE-ENVIRONMENT INTERACTION FOR FORAGE YIELD OF VETCH (VICIA SATIVA L.) IN MEDITERRANEAN ENVIRONMENTS. G. Pacucci and C.
GENOTYPE-ENVIRONMENT INTERACTION FOR FORAGE YIELD OF VETCH (VICIA SATIVA L.) IN MEDITERRANEAN ENVIRONMENTS ID # 12-16 G. Pacucci and C. Troccoli Department of Scienze delle Produzioni Vegetali, Università
More informationStatistical Techniques II EXST7015 Simple Linear Regression
Statistical Techniques II EXST7015 Simple Linear Regression 03a_SLR 1 Y - the dependent variable 35 30 25 The objective Given points plotted on two coordinates, Y and X, find the best line to fit the data.
More informationThe legacy of Sir Ronald A. Fisher. Fisher s three fundamental principles: local control, replication, and randomization.
1 Chapter 1: Research Design Principles The legacy of Sir Ronald A. Fisher. Fisher s three fundamental principles: local control, replication, and randomization. 2 Chapter 2: Completely Randomized Design
More informationG E INTERACTION USING JMP: AN OVERVIEW
G E INTERACTION USING JMP: AN OVERVIEW Sukanta Dash I.A.S.R.I., Library Avenue, New Delhi-110012 sukanta@iasri.res.in 1. Introduction Genotype Environment interaction (G E) is a common phenomenon in agricultural
More informationX -1 -balance of some partially balanced experimental designs with particular emphasis on block and row-column designs
DOI:.55/bile-5- Biometrical Letters Vol. 5 (5), No., - X - -balance of some partially balanced experimental designs with particular emphasis on block and row-column designs Ryszard Walkowiak Department
More informationStat 579: Generalized Linear Models and Extensions
Stat 579: Generalized Linear Models and Extensions Mixed models Yan Lu Feb, 2018, week 7 1 / 17 Some commonly used experimental designs related to mixed models Two way or three way random/mixed effects
More informationDESIGNING EXPERIMENTS AND ANALYZING DATA A Model Comparison Perspective
DESIGNING EXPERIMENTS AND ANALYZING DATA A Model Comparison Perspective Second Edition Scott E. Maxwell Uniuersity of Notre Dame Harold D. Delaney Uniuersity of New Mexico J,t{,.?; LAWRENCE ERLBAUM ASSOCIATES,
More informationConstruction of Partially Balanced Incomplete Block Designs
International Journal of Statistics and Systems ISS 0973-675 Volume, umber (06), pp. 67-76 Research India Publications http://www.ripublication.com Construction of Partially Balanced Incomplete Block Designs
More informationTwo-Way Factorial Designs
81-86 Two-Way Factorial Designs Yibi Huang 81-86 Two-Way Factorial Designs Chapter 8A - 1 Problem 81 Sprouting Barley (p166 in Oehlert) Brewer s malt is produced from germinating barley, so brewers like
More informationBlocks are formed by grouping EUs in what way? How are experimental units randomized to treatments?
VI. Incomplete Block Designs A. Introduction What is the purpose of block designs? Blocks are formed by grouping EUs in what way? How are experimental units randomized to treatments? 550 What if we have
More informationFactorial designs. Experiments
Chapter 5: Factorial designs Petter Mostad mostad@chalmers.se Experiments Actively making changes and observing the result, to find causal relationships. Many types of experimental plans Measuring response
More informationConstruction of lattice designs using MOLS
Malaya Journal of Matematik S(1)(2013) 11 16 Construction of lattice designs using MOLS Dr.R. Jaisankar a, and M. Pachamuthu b, a Department of Statistics, Bharathiar University, Coimbatore-641046, India.
More informationLarge Sample Properties of Estimators in the Classical Linear Regression Model
Large Sample Properties of Estimators in the Classical Linear Regression Model 7 October 004 A. Statement of the classical linear regression model The classical linear regression model can be written in
More informationDESAIN EKSPERIMEN Analysis of Variances (ANOVA) Semester Genap 2017/2018 Jurusan Teknik Industri Universitas Brawijaya
DESAIN EKSPERIMEN Analysis of Variances (ANOVA) Semester Jurusan Teknik Industri Universitas Brawijaya Outline Introduction The Analysis of Variance Models for the Data Post-ANOVA Comparison of Means Sample
More informationSTAT22200 Spring 2014 Chapter 8A
STAT22200 Spring 2014 Chapter 8A Yibi Huang May 13, 2014 81-86 Two-Way Factorial Designs Chapter 8A - 1 Problem 81 Sprouting Barley (p166 in Oehlert) Brewer s malt is produced from germinating barley,
More informationBLOCK DESIGNS WITH FACTORIAL STRUCTURE
BLOCK DESIGNS WITH ACTORIAL STRUCTURE V.K. Gupta and Rajender Parsad I.A.S.R.I., Library Avenue, New Delhi - 0 0 The purpose of this talk is to expose the participants to the interesting work on block
More informationAnalysis of Variance and Design of Experiments-II
Analysis of Variance and Design of Experiments-II MODULE - II LECTURE - BALANCED INCOMPLETE BLOCK DESIGN (BIBD) Dr. Shalabh Department of Mathematics & Statistics Indian Institute of Technology Kanpur
More informationCONSTRUCTION OF BLOCK DESIGNS WITH NESTED ROWS AND COLUMNS
Journal of Research (Science), Bahauddin Zakariya University, Multan, Pakistan. Vol. 8, No., July 7, pp. 67-76 ISSN - CONSTRUCTION OF BLOCK DESIGNS WITH NESTED ROWS AND COLUMNS L. Rasul and I. Iqbal Department
More informationMIXED MODELS THE GENERAL MIXED MODEL
MIXED MODELS This chapter introduces best linear unbiased prediction (BLUP), a general method for predicting random effects, while Chapter 27 is concerned with the estimation of variances by restricted
More informationConstruction of PBIBD (2) Designs Using MOLS
Intern. J. Fuzzy Mathematical Archive Vol. 3, 013, 4-49 ISSN: 30 34 (P), 30 350 (online) Published on 4 December 013 www.researchmathsci.org International Journal of R.Jaisankar 1 and M.Pachamuthu 1 Department
More information1 One-way Analysis of Variance
1 One-way Analysis of Variance Suppose that a random sample of q individuals receives treatment T i, i = 1,,... p. Let Y ij be the response from the jth individual to be treated with the ith treatment
More informationTopic 4: Orthogonal Contrasts
Topic 4: Orthogonal Contrasts ANOVA is a useful and powerful tool to compare several treatment means. In comparing t treatments, the null hypothesis tested is that the t true means are all equal (H 0 :
More informationMath 423/533: The Main Theoretical Topics
Math 423/533: The Main Theoretical Topics Notation sample size n, data index i number of predictors, p (p = 2 for simple linear regression) y i : response for individual i x i = (x i1,..., x ip ) (1 p)
More informationBIOL Biometry LAB 6 - SINGLE FACTOR ANOVA and MULTIPLE COMPARISON PROCEDURES
BIOL 458 - Biometry LAB 6 - SINGLE FACTOR ANOVA and MULTIPLE COMPARISON PROCEDURES PART 1: INTRODUCTION TO ANOVA Purpose of ANOVA Analysis of Variance (ANOVA) is an extremely useful statistical method
More informationCOMPLETELY RANDOM DESIGN (CRD) -Design can be used when experimental units are essentially homogeneous.
COMPLETELY RANDOM DESIGN (CRD) Description of the Design -Simplest design to use. -Design can be used when experimental units are essentially homogeneous. -Because of the homogeneity requirement, it may
More informationFractional Factorial Designs
k-p Fractional Factorial Designs Fractional Factorial Designs If we have 7 factors, a 7 factorial design will require 8 experiments How much information can we obtain from fewer experiments, e.g. 7-4 =
More informationOne-Way Analysis of Variance. With regression, we related two quantitative, typically continuous variables.
One-Way Analysis of Variance With regression, we related two quantitative, typically continuous variables. Often we wish to relate a quantitative response variable with a qualitative (or simply discrete)
More informationTopic 9: Factorial treatment structures. Introduction. Terminology. Example of a 2x2 factorial
Topic 9: Factorial treatment structures Introduction A common objective in research is to investigate the effect of each of a number of variables, or factors, on some response variable. In earlier times,
More information21.0 Two-Factor Designs
21.0 Two-Factor Designs Answer Questions 1 RCBD Concrete Example Two-Way ANOVA Popcorn Example 21.4 RCBD 2 The Randomized Complete Block Design is also known as the two-way ANOVA without interaction. A
More informationUnit 12: Analysis of Single Factor Experiments
Unit 12: Analysis of Single Factor Experiments Statistics 571: Statistical Methods Ramón V. León 7/16/2004 Unit 12 - Stat 571 - Ramón V. León 1 Introduction Chapter 8: How to compare two treatments. Chapter
More informationLinear models and their mathematical foundations: Simple linear regression
Linear models and their mathematical foundations: Simple linear regression Steffen Unkel Department of Medical Statistics University Medical Center Göttingen, Germany Winter term 2018/19 1/21 Introduction
More informationNature vs. nurture? Lecture 18 - Regression: Inference, Outliers, and Intervals. Regression Output. Conditions for inference.
Understanding regression output from software Nature vs. nurture? Lecture 18 - Regression: Inference, Outliers, and Intervals In 1966 Cyril Burt published a paper called The genetic determination of differences
More informationCOMBINING ABILITY ANALYSIS FOR CURED LEAF YIELD AND ITS COMPONENT TRAITS IN BIDI TOBACCO (NicotianatabacumL.)
International Journal of Science, Environment and Technology, Vol. 5, No 3, 2016, 1373 1380 ISSN 2278-3687 (O) 2277-663X (P) COMBINING ABILITY ANALYSIS FOR CURED LEAF YIELD AND ITS COMPONENT TRAITS IN
More informationAllow the investigation of the effects of a number of variables on some response
Lecture 12 Topic 9: Factorial treatment structures (Part I) Factorial experiments Allow the investigation of the effects of a number of variables on some response in a highly efficient manner, and in a
More informationChap The McGraw-Hill Companies, Inc. All rights reserved.
11 pter11 Chap Analysis of Variance Overview of ANOVA Multiple Comparisons Tests for Homogeneity of Variances Two-Factor ANOVA Without Replication General Linear Model Experimental Design: An Overview
More informationAn Old Research Question
ANOVA An Old Research Question The impact of TV on high-school grade Watch or not watch Two groups The impact of TV hours on high-school grade Exactly how much TV watching would make difference Multiple
More informationAnalysis of Variance
Statistical Techniques II EXST7015 Analysis of Variance 15a_ANOVA_Introduction 1 Design The simplest model for Analysis of Variance (ANOVA) is the CRD, the Completely Randomized Design This model is also
More informationOPTIMAL CONTROLLED SAMPLING DESIGNS
OPTIMAL CONTROLLED SAMPLING DESIGNS Rajender Parsad and V.K. Gupta I.A.S.R.I., Library Avenue, New Delhi 002 rajender@iasri.res.in. Introduction Consider a situation, where it is desired to conduct a sample
More informationSociology 6Z03 Review II
Sociology 6Z03 Review II John Fox McMaster University Fall 2016 John Fox (McMaster University) Sociology 6Z03 Review II Fall 2016 1 / 35 Outline: Review II Probability Part I Sampling Distributions Probability
More informationREGRESSION DIAGNOSTICS AND REMEDIAL MEASURES
REGRESSION DIAGNOSTICS AND REMEDIAL MEASURES Lalmohan Bhar I.A.S.R.I., Library Avenue, Pusa, New Delhi 110 01 lmbhar@iasri.res.in 1. Introduction Regression analysis is a statistical methodology that utilizes
More informationInference for Regression Simple Linear Regression
Inference for Regression Simple Linear Regression IPS Chapter 10.1 2009 W.H. Freeman and Company Objectives (IPS Chapter 10.1) Simple linear regression p Statistical model for linear regression p Estimating
More informationCOMBINED ANALYSIS OF DATA
COMBINED ANALYSIS OF DATA Krishan Lal I.A.S.R.I., Library Avenue, New Delhi- 110 01 klkalra@iasri.res.in 1. Introduction In large-scale experimental programmes it is necessary to repeat the trial of a
More informationthe experiment; and the distribution of residuals (or non-heritable
SOME GENOTYPIC FREQUENCIES AND VARIANCE COMPONENTS OCCURRING IN BIOMETRICAL GENETICS J. A. NELDER National Vegetable Research Station, Wellesbourne, Warwicks. Received 30.xi.5 i 1. INTRODUCTION MATHER
More informationMultiple Linear Regression
Multiple Linear Regression Simple linear regression tries to fit a simple line between two variables Y and X. If X is linearly related to Y this explains some of the variability in Y. In most cases, there
More informationB. L. Raktoe* and W. T. Federer University of Guelph and Cornell University. Abstract
BALANCED OPTIMAL SA'IURATED MAIN EFFECT PLANS OF 'IHE 2n FACTORIAL AND THEIR RELATION TO (v,k,'x.) CONFIGURATIONS BU-406-M by January, 1S72 B. L. Raktoe* and W. T. Federer University of Guelph and Cornell
More informationNON-PARAMETRIC METHODS IN ANALYSIS OF EXPERIMENTAL DATA
NON-PARAMETRIC METHODS IN ANALYSIS OF EXPERIMENTAL DATA Rajender Parsad I.A.S.R.I., Library Aenue, New Delhi 0 0. Introduction In conentional setup the analysis of experimental data is based on the assumptions
More informationAnalysis of Variance and Design of Experiments-I
Analysis of Variance and Design of Experiments-I MODULE VIII LECTURE - 35 ANALYSIS OF VARIANCE IN RANDOM-EFFECTS MODEL AND MIXED-EFFECTS MODEL Dr. Shalabh Department of Mathematics and Statistics Indian
More informationMixed-Model Estimation of genetic variances. Bruce Walsh lecture notes Uppsala EQG 2012 course version 28 Jan 2012
Mixed-Model Estimation of genetic variances Bruce Walsh lecture notes Uppsala EQG 01 course version 8 Jan 01 Estimation of Var(A) and Breeding Values in General Pedigrees The above designs (ANOVA, P-O
More informationDesign and Analysis of Experiments Prof. Jhareswar Maiti Department of Industrial and Systems Engineering Indian Institute of Technology, Kharagpur
Design and Analysis of Experiments Prof. Jhareswar Maiti Department of Industrial and Systems Engineering Indian Institute of Technology, Kharagpur Lecture - 27 Randomized Complete Block Design (RCBD):
More informationAlternative implementations of Monte Carlo EM algorithms for likelihood inferences
Genet. Sel. Evol. 33 001) 443 45 443 INRA, EDP Sciences, 001 Alternative implementations of Monte Carlo EM algorithms for likelihood inferences Louis Alberto GARCÍA-CORTÉS a, Daniel SORENSEN b, Note a
More informationIntroduction. Chapter 8
Chapter 8 Introduction In general, a researcher wants to compare one treatment against another. The analysis of variance (ANOVA) is a general test for comparing treatment means. When the null hypothesis
More informationChapter 4: Randomized Blocks and Latin Squares
Chapter 4: Randomized Blocks and Latin Squares 1 Design of Engineering Experiments The Blocking Principle Blocking and nuisance factors The randomized complete block design or the RCBD Extension of the
More informationto be tested with great accuracy. The contrast between this state
STATISTICAL MODELS IN BIOMETRICAL GENETICS J. A. NELDER National Vegetable Research Station, Wellesbourne, Warwick Received I.X.52 I. INTRODUCTION THE statistical models belonging to the analysis of discontinuous
More informationInference for the Regression Coefficient
Inference for the Regression Coefficient Recall, b 0 and b 1 are the estimates of the slope β 1 and intercept β 0 of population regression line. We can shows that b 0 and b 1 are the unbiased estimates
More informationDefinition 2.3. We define addition and multiplication of matrices as follows.
14 Chapter 2 Matrices In this chapter, we review matrix algebra from Linear Algebra I, consider row and column operations on matrices, and define the rank of a matrix. Along the way prove that the row
More informationHypothesis Testing hypothesis testing approach
Hypothesis Testing In this case, we d be trying to form an inference about that neighborhood: Do people there shop more often those people who are members of the larger population To ascertain this, we
More informationAMS 315/576 Lecture Notes. Chapter 11. Simple Linear Regression
AMS 315/576 Lecture Notes Chapter 11. Simple Linear Regression 11.1 Motivation A restaurant opening on a reservations-only basis would like to use the number of advance reservations x to predict the number
More informationpsyc3010 lecture 2 factorial between-ps ANOVA I: omnibus tests
psyc3010 lecture 2 factorial between-ps ANOVA I: omnibus tests last lecture: introduction to factorial designs next lecture: factorial between-ps ANOVA II: (effect sizes and follow-up tests) 1 general
More informationMean Comparisons PLANNED F TESTS
Mean Comparisons F-tests provide information on significance of treatment effects, but no information on what the treatment effects are. Comparisons of treatment means provide information on what the treatment
More informationMa 3/103: Lecture 25 Linear Regression II: Hypothesis Testing and ANOVA
Ma 3/103: Lecture 25 Linear Regression II: Hypothesis Testing and ANOVA March 6, 2017 KC Border Linear Regression II March 6, 2017 1 / 44 1 OLS estimator 2 Restricted regression 3 Errors in variables 4
More informationDo not copy, post, or distribute
14 CORRELATION ANALYSIS AND LINEAR REGRESSION Assessing the Covariability of Two Quantitative Properties 14.0 LEARNING OBJECTIVES In this chapter, we discuss two related techniques for assessing a possible
More informationExperimental Design and Data Analysis for Biologists
Experimental Design and Data Analysis for Biologists Gerry P. Quinn Monash University Michael J. Keough University of Melbourne CAMBRIDGE UNIVERSITY PRESS Contents Preface page xv I I Introduction 1 1.1
More informationLINEAR REGRESSION ANALYSIS. MODULE XVI Lecture Exercises
LINEAR REGRESSION ANALYSIS MODULE XVI Lecture - 44 Exercises Dr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur Exercise 1 The following data has been obtained on
More informationLecture 2: Linear Models. Bruce Walsh lecture notes Seattle SISG -Mixed Model Course version 23 June 2011
Lecture 2: Linear Models Bruce Walsh lecture notes Seattle SISG -Mixed Model Course version 23 June 2011 1 Quick Review of the Major Points The general linear model can be written as y = X! + e y = vector
More informationEstimation in Generalized Linear Models with Heterogeneous Random Effects. Woncheol Jang Johan Lim. May 19, 2004
Estimation in Generalized Linear Models with Heterogeneous Random Effects Woncheol Jang Johan Lim May 19, 2004 Abstract The penalized quasi-likelihood (PQL) approach is the most common estimation procedure
More informationLecture 4. Basic Designs for Estimation of Genetic Parameters
Lecture 4 Basic Designs for Estimation of Genetic Parameters Bruce Walsh. Aug 003. Nordic Summer Course Heritability The reason for our focus, indeed obsession, on the heritability is that it determines
More informationMultiple Regression. Inference for Multiple Regression and A Case Study. IPS Chapters 11.1 and W.H. Freeman and Company
Multiple Regression Inference for Multiple Regression and A Case Study IPS Chapters 11.1 and 11.2 2009 W.H. Freeman and Company Objectives (IPS Chapters 11.1 and 11.2) Multiple regression Data for multiple
More informationAbstract. Designs for variety trials with very low replication. How do we allow for variation between the plots? Variety Testing
Abstract Designs for variety trials with very low replication R A Bailey University of St Andrews QMUL (emerita) TU Dortmund, June 05 In the early stages of testing new varieties, it is common that there
More informationREGRESSION METHODS FOR STUDYING GENOTYPE- ENVIRONMENT INTERACTIONS. R. C. HARDWICK and J. T. WOOD
REGRESSION METHODS FOR STUDYING GENOTYPE- ENVIRONMENT INTERACTIONS R. C. HARDWICK and J. T. WOOD National Vegetable Research Station, Wellesbourne, Warwick Received 20.vii.71 1. INTRODUCTION WHEN the performance
More informationTentative solutions TMA4255 Applied Statistics 16 May, 2015
Norwegian University of Science and Technology Department of Mathematical Sciences Page of 9 Tentative solutions TMA455 Applied Statistics 6 May, 05 Problem Manufacturer of fertilizers a) Are these independent
More informationAnalysis of Variance. Read Chapter 14 and Sections to review one-way ANOVA.
Analysis of Variance Read Chapter 14 and Sections 15.1-15.2 to review one-way ANOVA. Design of an experiment the process of planning an experiment to insure that an appropriate analysis is possible. Some
More information14 Multiple Linear Regression
B.Sc./Cert./M.Sc. Qualif. - Statistics: Theory and Practice 14 Multiple Linear Regression 14.1 The multiple linear regression model In simple linear regression, the response variable y is expressed in
More information44.2. Two-Way Analysis of Variance. Introduction. Prerequisites. Learning Outcomes
Two-Way Analysis of Variance 44 Introduction In the one-way analysis of variance (Section 441) we consider the effect of one factor on the values taken by a variable Very often, in engineering investigations,
More informationFactorial Treatment Structure: Part I. Lukas Meier, Seminar für Statistik
Factorial Treatment Structure: Part I Lukas Meier, Seminar für Statistik Factorial Treatment Structure So far (in CRD), the treatments had no structure. So called factorial treatment structure exists if
More informationDr. Junchao Xia Center of Biophysics and Computational Biology. Fall /1/2016 1/46
BIO5312 Biostatistics Lecture 10:Regression and Correlation Methods Dr. Junchao Xia Center of Biophysics and Computational Biology Fall 2016 11/1/2016 1/46 Outline In this lecture, we will discuss topics
More informationDesign & Analysis of Experiments 7E 2009 Montgomery
1 What If There Are More Than Two Factor Levels? The t-test does not directly apply ppy There are lots of practical situations where there are either more than two levels of interest, or there are several
More informationChapter 1. Linear Regression with One Predictor Variable
Chapter 1. Linear Regression with One Predictor Variable 1.1 Statistical Relation Between Two Variables To motivate statistical relationships, let us consider a mathematical relation between two mathematical
More informationLecture 5: BLUP (Best Linear Unbiased Predictors) of genetic values. Bruce Walsh lecture notes Tucson Winter Institute 9-11 Jan 2013
Lecture 5: BLUP (Best Linear Unbiased Predictors) of genetic values Bruce Walsh lecture notes Tucson Winter Institute 9-11 Jan 013 1 Estimation of Var(A) and Breeding Values in General Pedigrees The classic
More informationGenetic Analysis for Heterotic Traits in Bread Wheat (Triticum aestivum L.) Using Six Parameters Model
International Journal of Current Microbiology and Applied Sciences ISSN: 2319-7706 Volume 7 Number 06 (2018) Journal homepage: http://www.ijcmas.com Original Research Article https://doi.org/10.20546/ijcmas.2018.706.029
More informationRECENT DEVELOPMENTS IN VARIANCE COMPONENT ESTIMATION
Libraries Conference on Applied Statistics in Agriculture 1989-1st Annual Conference Proceedings RECENT DEVELOPMENTS IN VARIANCE COMPONENT ESTIMATION R. R. Hocking Follow this and additional works at:
More informationFormal Statement of Simple Linear Regression Model
Formal Statement of Simple Linear Regression Model Y i = β 0 + β 1 X i + ɛ i Y i value of the response variable in the i th trial β 0 and β 1 are parameters X i is a known constant, the value of the predictor
More informationSolutions to Final STAT 421, Fall 2008
Solutions to Final STAT 421, Fall 2008 Fritz Scholz 1. (8) Two treatments A and B were randomly assigned to 8 subjects (4 subjects to each treatment) with the following responses: 0, 1, 3, 6 and 5, 7,
More informationStats fest Analysis of variance. Single factor ANOVA. Aims. Single factor ANOVA. Data
1 Stats fest 2007 Analysis of variance murray.logan@sci.monash.edu.au Single factor ANOVA 2 Aims Description Investigate differences between population means Explanation How much of the variation in response
More information