Scribd Logo
Upload

Welcome to Scribd!

  • 24 views

    Notes 2

    Uploaded by

    Vaibhav Sehgal
    The document describes using linear regression to estimate salary based on percentage scored in 10th grade. It finds that percentage in 10th grade significantly impacts salary, with each 1% increase in 10th grade percentage associated with a Rs. 3076 increase in average salary. It provides equations to estimate salary for people who scored 80% and 40% in 10th grade. A second linear regression models treatment cost based on age but finds no significant relationship between age and cost.

    Copyright:

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

    Notes 2

    Uploaded by

    Vaibhav Sehgal
    24 views7 pages
    The document describes using linear regression to estimate salary based on percentage scored in 10th grade. It finds that percentage in 10th grade significantly impacts salary, with each 1% increase in 10th grade percentage associated with a Rs. 3076 increase in average salary. It provides equations to estimate salary for people who scored 80% and 40% in 10th grade. A second linear regression models treatment cost based on age but finds no significant relationship between age and cost.

    Original Title

    Notes2

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document describes using linear regression to estimate salary based on percentage scored in 10th grade. It finds that percentage in 10th grade significantly impacts salary, with each 1% increase in 10th grade percentage associated with a Rs. 3076 increase in average salary. It provides equations to estimate salary for people who scored 80% and 40% in 10th grade. A second linear regression models treatment cost based on age but finds no significant relationship between age and cost.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    24 views7 pages

    Notes 2

    Uploaded by

    Vaibhav Sehgal
    The document describes using linear regression to estimate salary based on percentage scored in 10th grade. It finds that percentage in 10th grade significantly impacts salary, with each 1% increase in 10th grade percentage associated with a Rs. 3076 increase in average salary. It provides equations to estimate salary for people who scored 80% and 40% in 10th grade. A second linear regression models treatment cost based on age but finds no significant relationship between age and cost.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    of 7

    Determine an equation which estimates salary?

    Coefficientsa
    Standardized
    Unstandardized Coefficients Coefficients
    Model B Std. Error Beta t Sig.
    1 (Constant) 61555.355 66701.901 .923 .361
    Percentage in grade 10 3076.177 1031.526 .395 2.982 .004
    a. Dependent Variable: Salary

    Salary (Y) = 61555.355 + 3076.177*Percentage in grade 10

    What would be the salary of MBA graduate if percentage in grade 10 is 80%?

    Salary = 61555.355 + 3076.177*80 = 307649.515

    What would be the salary of MBA graduate if percentage in grade 10 is 40%?

    Salary = 61555.355 + 3076.177*40 = 184602.435

    Interpretation:

    For every 1 percentage increase in grade 10, the salary of the MBA student increase at the rate of
    3076.177 on an average.

    1. Independent variable is to be statistically significant -> Concept of Verifying Regression


    coefficient
    H0: There is no relationship between X (Percentage in grade 10) and Y(Salary)
    H1: There is a relationship between X (Percentage in grade 10) and Y(Salary)
    H0: β1=0
    H1: β1 ≠ 0
    P-value < 0.05, reject Null Hypothesis, which means, there is a relationship between
    X (Percentage in grade 10) and Y(Salary)

    2. Overall Equation is statistically significant.

    Salary (Y) = 61555.355 + 3076.177*Percentage in grade 10


    H0: There is no statistically significant relationship between Y(Salary) and Percentage any of
    the explanatory variables.
    H1: Not all regression coefficients are zero
    ANOVAa
    Model Sum of Squares df Mean Square F Sig.
    1 Regression 45078257910.04 1 45078257910.04 8.893 .004b
    8 8
    Residual 243302058889.9 48 5068792893.541
    52
    Total 288380316800.0 49
    00
    a. Dependent Variable: Salary
    b. Predictors: (Constant), Percentage in grade 10

    Since the p-value is less than α, Reject H0, which means that, the regression coefficient is not equal
    to zero and the overall model is significant.

    Coefficient of Determination(R-Square)

    Salary (Y) = 61555.355 + 3076.177*Percentage in grade 10

    Variation explained by independent variable to determine the change in dependent variable.

    Model Summary
    Adjusted R Std. Error of the
    Model R R Square Square Estimate
    1 .395a .156 .139 71195.456
    a. Predictors: (Constant), Percentage in grade 10

    R-Square = 0.156 = 15.6% variation in Salary is determined by Percentage in grade 10

    Square root of R

    Fulfilling the assumptions of simple Linear Regression:

    1. Linearity (Yes)
    2. Homoscedasticity (this will fulfil when the scatterplot doesn’t take funnel shape)
    3. Independence (Yes)
    4. Normality

    Predicted values

    Residual values – actual values


    Independent variable: Age

    Dependent variable: Cost of treatment

    Coefficientsa
    Standardized
    Unstandardized Coefficients Coefficients
    Model B Std. Error Beta t Sig.
    1 (Constant) 189010.378 31067.222 6.084 .000
    Age 323.642 778.396 .067 .416 .680
    a. Dependent Variable: CostofTreatment

    Cost of Treatment (Y) = 189010.378 + 323.642*Age

    1. Independent variable is to be statistically significant -> Concept of Verifying Regression


    coefficient
    H0: There is no relationship between X (Age) and Y (Cost of Treatment)
    H1: There is a relationship between X (Age) and Y (Cost of Treatment)
    H0: β1=0
    H1: β1 ≠ 0
    P-value > 0.05, accept Null Hypothesis, which means, there is no relationship between
    X (Age) and Y (Cost of Treatment)

    2. Overall Equation is statistically significant.

    Cost of Treatment (Y) = 189010.378 + 323.642*Age

    H0: There is no statistically significant relationship between Y (Cost of treatment) and age
    any of the explanatory variables.
    H1: Not all regression coefficients are zero

    ANOVAa
    Model Sum of Squares df Mean Square F Sig.
    1 Regression 1535787003.152 1 1535787003.152 .173 .680b
    Residual 337587122408.6 38 8883871642.332
    32
    Total 339122909411.7 39
    84
    a. Dependent Variable: CostofTreatment
    b. Predictors: (Constant), Age

    Since the p-value is greater than α, Accept H0, which means that, the regression coefficient is equal
    to zero and the overall model is not significant.

    Coefficient of Determination(R-Square)

    Cost of Treatment (Y) = 189010.378 + 323.642*Age

    Variation explained by independent variable to determine the change in dependent variable.

    Model Summary
    Adjusted R Std. Error of the
    Model R R Square Square Estimate
    1 .067a .005 -.022 94254.29243452
    218000
    a. Predictors: (Constant), Age

    R-Square = 0.005 = 0.5% variation in Cost of Treatment is determined by Age

    You might also like