Scribd Logo
Upload

Welcome to Scribd!

  • 2 views

    Experiment-3: Nisheeta Thakur 121A1114 D2

    Uploaded by

    Nisheeta Thakur
    To implement multivariate linear regression in Python.

    Copyright:

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

    Experiment-3: Nisheeta Thakur 121A1114 D2

    Uploaded by

    Nisheeta Thakur
    2 views3 pages
    To implement multivariate linear regression in Python.

    Original Title

    121A1114_D2_ML_EXP3

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    To implement multivariate linear regression in Python.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    2 views3 pages

    Experiment-3: Nisheeta Thakur 121A1114 D2

    Uploaded by

    Nisheeta Thakur
    To implement multivariate linear regression in Python.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    of 3

    NISHEETA THAKUR

    121A1114 D2
    Experiment-3

    Aim: To implement multivariate linear regression in Python.

    To implement Multivariate Linear Regression in Python and understand the


    concept using:

    1. One independent variable and multiple dependent variables.


    2. Multiple independent variables and multiple dependent variables,
    utilizing the matrix method for parameter estimation.

    Theory:
    Multivariate Linear Regression:
    Multivariate Linear Regression refers to regression involving multiple dependent
    variables (DV). This contrasts with simple multiple regression, where there is only
    one dependent variable.
    The equation for multivariate linear regression can be generalized as:
    Y=Xβ+ϵ
    where:
     Y: n×m matrix of dependent variables (responses), with m being the number
    of dependent variables.
     X: n×(p+1) matrix of independent variables (including a column of 1s for
    the intercept).
     β: (p+1)×m matrix of regression coefficients.
     ϵ: n×m matrix representing the error or residuals.
    The optimal parameters β can be estimated using the normal equation:

    This method minimizes the sum of squared errors to find the best-fitting line(s).
    Part 1: One Independent Variable and Multiple Dependent Variables (1
    Independent Variable, Multiple Dependent Variables)
    Case Explanation:
    In this case, we have one independent variable XXX and multiple dependent
    variables Y1, Y2,…,Ym. The goal is to find the regression coefficients for each
    dependent variable with respect to the single independent variable.
    The equation becomes:

    Using the matrix form, we compute:

    where X is a column matrix of the independent variable, and Y is a matrix of the


    dependent variables.

    Part 2: Multiple Independent Variables and Multiple Dependent Variables


    (Multiple Independent Variables, Multiple Dependent Variables)
    Case Explanation:
    In this case, we have multiple independent variables and multiple dependent
    variables. The goal is to find the optimal coefficients that minimize the residual error
    for each dependent variable.
    The equation becomes:

    In matrix form, the optimal parameter matrix β hat is again computed as:

    Here, both X and Y are matrices with multiple columns representing multiple
    predictors and responses, respectively.
    Program :

    Output :

    Conclusion:

     Multivariate Linear Regression allows us to model the relationships between


    multiple independent and multiple dependent variables.
     Using the matrix method, we can efficiently compute the optimal parameters
    by solving the normal equation.
     One independent variable with multiple dependent variables captures
    simpler relationships, while multiple independent and dependent variables
    provide a more comprehensive model.

    You might also like