Econometric Presentation

download Econometric Presentation

of 14

Transcript of Econometric Presentation

  • 8/2/2019 Econometric Presentation

    1/14

    Click to edit Master subtitle style

    4/11/12

    THE RELATIONSHIP BETWEEN GDPTOWARDS EMPLOYMENT AND

    UNEMPLOYMENT

    By: Bertilla Voyn Bj Bius (BG10110035)Kamsianah Bt Tajudin (BG10110164)Nursyuhada Binti Muhamad Wazir (BG10110388)

  • 8/2/2019 Econometric Presentation

    2/14

    4/11/12

    INTRODUCTION

  • 8/2/2019 Econometric Presentation

    3/14

    4/11/12

    THEORETICAL FRAMEWORK

  • 8/2/2019 Econometric Presentation

    4/14

    4/11/12

    LITERATURE REVIEW

  • 8/2/2019 Econometric Presentation

    5/14

    4/11/12

    METHODOLOGY AND DATA

    Simple Regression

    Multiple Regression

    Log-linear Regression Model

    Normality Distribution Heteroscedasticity test (4step)

    Autocorrelation test

    ( 2 step )

    The data is from

    The data have three variable which are GDP as dependent,Employment and Unemployment as independent

    The range is from 1959 to 2010 and it is quarterly.

  • 8/2/2019 Econometric Presentation

    6/14

    4/11/12

    MULTIPLE REGRESSION

    = -5674.722+0.125205X1-0.032600X2

    Se = (185.2340) (0.002238) (0.022553)

    t = (-30.63542) (55.94006) (-1.445458)

    r2 = 0.962382p = (0.0000) (0.0000) (0.1499)

    f-statistic= 2622.286 d = 0.011071

    Equation above shows, the estimated value of B1 is -5674.72,B2 is 0.12 and B3 is -0.033. Thus, if the employment goes up

    by one unit (1%), gross domestic product (GDP) is expected toincrease on the average by about 0.12 and when unemployment

    increase by one unit (1%), gross domestic product (GDP) isexpected to decrease on the average by about -0.033. The

  • 8/2/2019 Econometric Presentation

    7/14

    4/11/12

    NORMALITY TEST(HYPOTHESIS TESTING)

    Step 1: Hypothesis Testing

    H0: The error term is normally distributed

    H1: The error term is not normally distributed

    Step 2: Decision Rule

    Reject the null hypothesis if probability is less than alpha;otherwise do not reject the null hypothesis.

    Step 3 : Compute

    Probability = 0.016572

    alpha () = 0.05 (5%)

    Step 4 : Decision

    The probability0.016572 is less than alpha() 0.05. Therefore,we reject the null hypothesis.

  • 8/2/2019 Econometric Presentation

    8/14

    4/11/12

    F-STATISTIC(HYPOTHESIS TESTING)

    Step 1: Hypothesis Testing

    H0: R2=0

    H1: R20

    Step 2: Decision Rule

    Reject the null hypothesis when F-statistic is greater thancritical value (c.v)and it will shows that gdp has acorrelation between employed and unemployed, otherwise

    do not reject. Step 3: Compute

    R2=0.962382 F-statistic = 2262.26

    Numerator (n) = 2 d.f = n-(k+1) = 208-(2+1) = 205

  • 8/2/2019 Econometric Presentation

    9/14

    4/11/12

    RAMSEY TEST(HYPOTHESIS TESTING)

    STEP 1:

    H0: misspecification error

    H1: not misspecification error

    STEP 2:

    The alternative hypothesis contains statement that istrying to prove the error term is not misspecification error.When probability chi-square is less than alpha(), reject

    the null hypothesis and it shows that the error term ismisspecification error.

    STEP 3:

    Probability Chi-Square = 0.0000

    =

  • 8/2/2019 Econometric Presentation

    10/14

    4/11/12

    HETEROSCEDASTICITY(WHITE)

    Step 1 : Hypothesis Testing

    H0: The is no heteroskedasticity problem

    H1: The is heteroskedasticity problem

    Step 2: Decision Rule

    Reject the null hypothesis if the probability is less thanalpha and this ; otherwise do not reject the nullhypothesis.

    Step 3 : Compute

    Error term(r2) = 149.3701 where the probability Chi-Square is 0.0000

    alpha () = 0.05 (5%)

  • 8/2/2019 Econometric Presentation

    11/14

    4/11/12

    HETEREOSCEDASTICITY(WHITE)

  • 8/2/2019 Econometric Presentation

    12/14

    4/11/12

    AUTOCORRELATION(DURBIN-WATSON STATISTIC)

    Step 1 : Hypothesis

    TestingH0: There is no

    autocorrelation problemH1: There is

    autocorrelation problem

  • 8/2/2019 Econometric Presentation

    13/14

    4/11/12

    AUTOCORRELATION(NEWEY TEST)

  • 8/2/2019 Econometric Presentation

    14/14

    4/11/12

    DISCUSSIONANDCONCLUSION