Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 10 views

    LSM Method-6-1-2022

    Uploaded by

    proto plasm

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd

    LSM Method-6-1-2022

    Uploaded by

    proto plasm
    10 views5 pages

    Original Title

    LSM method-6-1-2022

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    10 views5 pages

    LSM Method-6-1-2022

    Uploaded by

    proto plasm

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    You are on page 1of 5

    Least-squares Methodology:

    Yij = µ + Si +eij

    Si = the effect of ith season (i=1, 2, 3, 4)

    µ S1 S2 S3 S4 RHM

    µ n. n1 n2 n3 n4 Y.
    S1 n 1 n1 - - - Y1
    S2 n2 - n2 - - Y2
    S3 n3 - - n3 - Y3
    S4 n4 - - - n4 Y4

    n. = ∑ni = n1 + n2 + n3 + n4
    Y. = ∑Yi = Y1 + Y2 + Y3 + Y4

    Properties of LS equations:
    1. LSE are symmetry
    2. Whatever the coefficients are found in µ equation, the
    same coefficients are found in diagonal elements of the
    symmetry matrix.
    3. Coefficient of µ i.e., coeffi (µ) =∑Si = S1 + S2 + S3 + S4
    µ in RHS i.e., Y. = ∑ RHS of all sub effect = ∑ Yi = Y1 + Y2 + Y3 +
    Y4
    4. All off-diagonal elements of corresponding sub-effect
    should be zero

    Imposing Restriction
    In the above case, we are not able to solve the above equations
    until some restrictions imposed to remove the empty cells
    carrying in the above matrix. This can be done by row and
    column wise reduction of the matrix except µ equation to
    obtained the reduced least squares equations and maintain the
    symmetry in the equation which is one of the properties of
    least squares equations.
    The estimated constants obtained by inverting the reduced
    least squares equations and multiplying the inverse elements
    with the reduced right hand members of the equation. In terms
    of notation, we can write

    µ m11 m12 m13 m14 Y.


    S1 m21 m22 m23 m24 Y 1 - Y4
    S2 = m31 m32 m33 m34 Y2 - Y4
    S3 m41 m42 m43 m44 Y3 - Y4

    Where, mij is the inverse elements of the reduced least-squares


    variance-covariance matrix

    Restriction - ∑Si = 0 , i.e., S1 + S2 + S3 + S4 = 0 because each and


    every observation in each subset is deviating from mean.
    Derived constant should be S4 = - (S1 + S2 + S3)

    Setting up ANOVA
    1. SS (Total) = SS of individual observation – CF
    = ∑ ∑Yij2 - G2/N
    Where, G = Grand Total and N is the total number of
    observation
    2. SS (Season) = B’Z-1B, where B’ = [S1 S2 S3], is the column
    vector of the set constants
    B = is the row vector, i.e., transpose of B’ and
    m22 m23 m24

    Z-1 = m32 m33 m34 = is the inverse of the

    m42 m43 m44


    segments of the variance-covariance matrix corresponding by
    row and column to the set of the constants, i.e., S1 S2 and S3.
    Y.
    3. SS(error) = = ∑ ∑Yij2 – [µ S1 S2 S3] Y1- Y4
    Y2 - Y4
    Y3 - Y4
    Thus the ANOVA will be as follows:

    SOV df SS MSS F value


    Between p-1
    Seasons B’Z-1B B’Z-1B/p-1 = MSA/MSE
    MSA

    Error n. - p δ e2 δe2 /n. –p =


    MSE

    The F- value is compared with table value for testing the


    significance of the factor.

    You might also like