Scribd Logo
Upload

Welcome to Scribd!

  • 22 views

    Tutorial 3 Ans

    Uploaded by

    kajani nesakumar
    1) The document provides sample data from a company over 12 years to analyze the relationship between year of operation (X) and number of employees (Y). 2) Regression analysis finds that for each additional year, the number of employees is expected to increase by 45.73, and in the first year there would be 153.6 employees. 3) Various hypothesis tests show that the slope is significantly greater than 30 employees per year, and the regression model fits the data well in explaining variation in employees.

    Copyright:

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

    Tutorial 3 Ans

    Uploaded by

    kajani nesakumar
    22 views4 pages
    1) The document provides sample data from a company over 12 years to analyze the relationship between year of operation (X) and number of employees (Y). 2) Regression analysis finds that for each additional year, the number of employees is expected to increase by 45.73, and in the first year there would be 153.6 employees. 3) Various hypothesis tests show that the slope is significantly greater than 30 employees per year, and the regression model fits the data well in explaining variation in employees.

    Copyright

    © © All Rights Reserved

    Available Formats

    DOCX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    1) The document provides sample data from a company over 12 years to analyze the relationship between year of operation (X) and number of employees (Y). 2) Regression analysis finds that for each additional year, the number of employees is expected to increase by 45.73, and in the first year there would be 153.6 employees. 3) Various hypothesis tests show that the slope is significantly greater than 30 employees per year, and the regression model fits the data well in explaining variation in employees.

    Copyright:

    © All Rights Reserved
    Download as DOCX, PDF, TXT or read online from Scribd
    Download as docx, pdf, or txt
    22 views4 pages

    Tutorial 3 Ans

    Uploaded by

    kajani nesakumar
    1) The document provides sample data from a company over 12 years to analyze the relationship between year of operation (X) and number of employees (Y). 2) Regression analysis finds that for each additional year, the number of employees is expected to increase by 45.73, and in the first year there would be 153.6 employees. 3) Various hypothesis tests show that the slope is significantly greater than 30 employees per year, and the regression model fits the data well in explaining variation in employees.

    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 4

    UMED8M-15-2 - Introductory Econometrics

    Tutorial 3: Hypothesis testing - Answers

    1.
    A company which has been operating for 12 years has experienced vary rapid growth. The
    company wish to predict the number of employees required in future years if growth continues at
    the same rate. Data is available for some statistical analysis.
    Assume that the relationship is of the form:

    Y^ i=B1 + B 2 X i i=1,..,12

    where X is year of operation and Y is the number of employees The disturbance terms, e i, are assumed
    to be normally and independently distributed with zero mean and constant variance σ . Preliminary
    2

    analysis of the sample data with 12 observations produces the following sample information:
    n n n
    ∑ x i y i=6539 ∑ x 2i =143 ∑ y 2i =343984
    1 1 1
    n
    X =6.5 Y=450 .83 ∑ e 2i =44973
    1
    n
    ∑ X 2i =650
    1
    Lower case letters indicate that the variables are measured as deviations from their respective
    sample means i.e xi = Xi - X .

    Use the above sample information to answer all the following questions. Show explicitly all calculations.
    a) Calculate estimates, b1 and b2, for the unknown coefficients, B1 and B2.

    Using the formulae for OLS (you are not expected to prove these) we have that, using deviations of the
    variables from their means:

    b) State the regression equation and interpret the estimated coefficients.

    b2: for each extra year, expect number of employees to increase by 45.73 on average.

    b1: at beginning of period expect number of employees to be 153.6

    Tutorial 3 - answers 1/4


    c) Calculate the estimated standard errors of the estimated coefficients.

    So se(b1) = √1703.5 = 41.27


    se(b2) = √31.45 = 5.61

    d) Perform a test to investigate if B2 is greater than 30 at the 5% significance level. Interpret this result.
    H0: B2 ≤ 30
    H1: B2 > 30

    The test is calculated as:

    Compare t with the critical value of t obtained from the Student Tables with 12-2 degrees of
    freedom.

    For a one-tailed test and 10 degrees of freedom,

    the 5% critical value is = 1.812.


    t > t* or 2.804 > 1.812 so we correspondingly reject the null in favour of the alternative.
    The slope coef is therefore significantly greater than 30, i.e.it is expected that the number of
    employees increases by more than 30 each year.

    e) Compute the 95% confidence interval for the intercept. Interpret this result.

    Degrees of freedom are n-2 = 10 so: = 2.228

    P(-2.228 < t < 2.228) = 0.95

    P(-2.228 < < 2.228) = 0.95

    i.e

    se(b1) = 41.27

    so CI

    Tutorial 3 - answers 2/4


    i.e. (61.66, 245.56)
    It is 95% confident that the true, unknown, intercept in the population regression line is between 61.66
    and 245.56. This is quite a wide interval but wholly above zero so the initial number of employees was
    at least 61.66 (with associated confidence)

    f) Compute the value of R2 for the estimated regression. Briefly explain what the estimated value of R2
    means.

    86.93% of variation in number of employees is explained by variation in year of operation. This is a very
    good fit.

    g) Perform a test of the overall fit of the model, i.e. a joint test for the significance of the estimated
    coefficients.

    H0: R2 = 0
    H1: R2 > 0

    df1 = 1; df2 = 10
    5% value for F1,10=4.96, 1% value for F1,6 = 10.00

    Reject H0: R2=0 in favour of alternative H1: R2>0 at 5% level because F= 66.51 > F1,10 =4.96 at 5%
    Can also reject H0: R2=0 in favour of alternative H1: R2>0 at 1% level because F= 66.51 > F1,10 =10.00 at
    1%
    Conclude that there is explanatory power in the model at both 1% and 5% significance levels.

    h) Predict the number of employees needed in year 14 if growth continues at the same rate. Calculate a
    99% prediction interval for the prediction.

    employees

    Se(Y0) =

    Critical value for 99% confidence interval d.f=10 is 3.1698

    Tutorial 3 - answers 3/4


    Confidence interval given by:

    LB = 793.826 - 3.1698*46.3 = 647.10

    UB = 793.826 + 3.1698*46.3 = 940.55

    Tutorial 3 - answers 4/4

    You might also like