lecture 14 econometrics

Upload: jomanous

Post on 14-Apr-2018

224 views

Category:

Documents


0 download

TRANSCRIPT

  • 7/29/2019 Lecture 14 econometrics

    1/35

    Lecture 15. Dummy variables, continued

    Seasonal effects in time series

    Consider relation between electricity consumption Y and electricityprice X .

    The data are quarterly time series.

    First model

    tttuXY ++= lnln 21

    What is the interpretation of 2 ?

  • 7/29/2019 Lecture 14 econometrics

    2/35

    Because electricity consumption depends on the weather and specialcircumstances (Christmas, summer holidays) we expect is to be

    different in the quarters, even if the price is constant.

    Solution: Define

    11 =tD ift is the first quarter of a year

    01=

    tD if not

    Define 432 ,, ttt DDD analogously for the other quarters.

  • 7/29/2019 Lecture 14 econometrics

    3/35

    To allow for differences in the average consumption betweenquarters we write

    44332211 DDD +++=

    Substitution in the regression model gives

    ttttttuXDDDY +++++= lnln 24433221

    Why not 1tD in model?

  • 7/29/2019 Lecture 14 econometrics

    4/35

    Average electricity consumption in the four quarters (given price tX )

    Quarter 1ttt

    XXYE ln)|(ln 21 +=

    Quarter 2ttt

    XXYE ln)|(ln 221 ++=

    Quarter 3ttt

    XXYE ln)|(ln 231 ++=

    Quarter 4ttt

    XXYE ln)|(ln 241 ++=

    Interpretation 432 ,, : Relative change (relative to quarter 1) of

    electricity consumption in quarters 2,3,4.

  • 7/29/2019 Lecture 14 econometrics

    5/35

    Change of reference quarter to quarter 2:

    44331211 DDD +++=

    Intercept in the four quarters

    Reference quarter is quarter 1

    4131211 ,,, +++

    Reference quarter is quarter 2

    4131121 ,,, +++

  • 7/29/2019 Lecture 14 econometrics

    6/35

    Hence

    24423322211 ,,, ===+=

    The same relations hold for the OLS estimates. Change of referencequarter does not require re-estimation.

    Same result holds for change in reference category for any

    qualitative variable with more than two values (e.g. earlier example

    with type of work)

  • 7/29/2019 Lecture 14 econometrics

    7/35

    If we want to investigate whether price elasticity depends on seasonwe write

    48372652 DDD +++=

    Substitution gives

    ttttttt

    ttttt

    uXDXDXD

    XDDDY

    +++

    +++++=

    lnlnln

    lnln

    483726

    54433221

  • 7/29/2019 Lecture 14 econometrics

    8/35

    Price elasticities in the four quarters

    Quarter 1 5

    Quarter 2 65 +

    Quarter 3 75 + Quarter 4 85 +

  • 7/29/2019 Lecture 14 econometrics

    9/35

    Tests:

    Average demand does not change with the season

    Price elasticity constant over seasons

    Derive price elasticities if we choose period 2 as reference period.

  • 7/29/2019 Lecture 14 econometrics

    10/35

    Structural change

    Events may change the relation between economic variables.

    Consider time series data on dependent variableY

    andindependent variable X for e.g. years nt ,,1K= .

    In year 0nt= some event happens.

    This event induces a structural change if the regression

    coefficients change due to the event.

  • 7/29/2019 Lecture 14 econometrics

    11/35

    Original model (no structural change)

    ntuXYttt

    ,,1, K=++=

    Model with structural change in0

    n :

    011 ,,1, ntuXY ttt K=++=

    nntuXYttt

    ,,1,)( 02121 K+=++++=

  • 7/29/2019 Lecture 14 econometrics

    12/35

    This is equivalent to introducing the dummy variable

    0=t

    D for 0,,1 nt K=

    1=

    t

    D

    fornnt

    ,,10K+=

    with the model

    ntuXDXDYtttttt

    ,,1,2121

    K=++++=

    Test for structural change can be done in two ways

    Estimate separate models and compare ESS

    Estimate model with dummy and test 0,0 22 ==

    This gives the same value for the test statistic.

  • 7/29/2019 Lecture 14 econometrics

    13/35

    Outliers

    There may be individual observations that do not fit the relation

    See output/graphs

    Reason:

    Omitted variables

    Error in the data

    Some unknown event/circumstance

    How to check this?

  • 7/29/2019 Lecture 14 econometrics

    14/35

    Introduce dummy variable

    123, =iD for observation 23 (and 0 otherwise)

    Include this in the regression model and test whether coefficient is 0.

    See output.

  • 7/29/2019 Lecture 14 econometrics

    15/35

    Dependent Variable: LNWAGEMethod: Least SquaresDate: 11/01/01 Time: 08:42

    Sample: 1 49Included observations: 49

    Variable Coefficient Std. Error t-Statistic Prob.

    C 6.864366 0.186127 36.88002 0.0000EDUC 0.052987 0.017107 3.097432 0.0034

    EXPER 0.020776 0.006321 3.286999 0.0020AGE -0.002250 0.003804 -0.591382 0.5574

    RACE 0.071479 0.081543 0.876575 0.3856GENDER 0.242610 0.071645 3.386300 0.0015

    R-squared 0.470916 Mean dependent var 7.454952Adjusted R-squared 0.409395 S.D. dependent var 0.312741S.E. of regression 0.240344 Akaike info criterion 0.100786Sum squared resid 2.483904 Schwarz criterion 0.332438Log likelihood 3.530733 F-statistic 7.654508Durbin-Watson stat 1.708658 Prob(F-statistic) 0.000032

  • 7/29/2019 Lecture 14 econometrics

    16/35

    -0.6

    -0.4

    -0.2

    0.0

    0.2

    0.4

    0.6

    5 10 15 20 25 30 35 40 45

    LNWAGE Residuals

  • 7/29/2019 Lecture 14 econometrics

    17/35

    obs Actual Fitted Residual Residual Plot

    1 7.20415 7.20983 -0.00568

    2 7.79770 7.64738 0.150323 7.44717 7.47824 -0.03107

    4 7.28688 7.44899 -0.16212

    5 7.40184 7.57869 -0.17685

    6 7.20415 7.27025 -0.06610

    7 7.37901 7.36391 0.01509

    8 7.04229 7.06440 -0.02211

    9 7.35628 7.78801 -0.43173

    10 7.31055 7.61880 -0.30825

    11 7.11802 7.16206 -0.0440412 7.20415 7.19236 0.01179

    13 7.20415 7.36341 -0.15926

    14 8.12829 7.90676 0.22153

    15 7.51698 7.67094 -0.15396

    16 6.88857 7.34406 -0.45549

    17 7.20415 7.54399 -0.33984

    18 7.35628 7.14941 0.20687

    19 7.07918 7.21830 -0.13911

    20 7.20415 7.41777 -0.21362

    21 7.20415 7.35979 -0.15564

    22 7.68110 7.56638 0.11472

    23 7.24566 7.69937 -0.45372

    24 7.65681 7.51358 0.14323

    25 7.70436 7.68690 0.01746

    26 8.18172 7.69434 0.48738

    27 7.58680 7.32344 0.26337

    28 7.11802 7.09935 0.01866

    29 7.56320 7.43966 0.12354

    30 7.68018 7.43267 0.2475131 7.76853 7.35286 0.41568

    32 7.20415 7.41537 -0.21122

    33 7.51698 7.28394 0.23304

    34 7.86825 7.65101 0.21724

    35 7.83716 7.72690 0.11026

    36 7.37901 7.39163 -0.01262

    37 7.51698 7.69670 -0.17973

    38 7.70436 7.45990 0.24446

    39 7.33237 7.18733 0.1450440 7.28688 7.29798 -0.01110

    41 8.10380 8.01117 0.09263

    42 8.25140 7.75389 0.49751

    43 7.51698 7.48605 0.03093

    44 7.28688 7.29015 -0.00328

    45 7.26753 7.58319 -0.31567

  • 7/29/2019 Lecture 14 econometrics

    18/35

    Dependent Variable: LNWAGEMethod: Least SquaresDate: 10/29/01 Time: 22:21

    Sample: 1 49Included observations: 49

    Variable Coefficient Std. Error t-Statistic Prob.

    C 6.789626 0.182398 37.22432 0.0000GENDER 0.261107 0.069438 3.760286 0.0005

    AGE -0.001271 0.003687 -0.344724 0.7320EXPER 0.018787 0.006149 3.055107 0.0039

    EDUC 0.061945 0.016981 3.647842 0.0007RACE 0.065118 0.078464 0.829904 0.4113D23 -0.530696 0.249938 -2.123314 0.0397

    R-squared 0.522205 Mean dependent var 7.454952Adjusted R-squared 0.453948 S.D. dependent var 0.312741S.E. of regression 0.231101 Akaike info criterion 0.039638Sum squared resid 2.243118 Schwarz criterion 0.309898Log likelihood 6.028867 F-statistic 7.650623Durbin-Watson stat 1.653329 Prob(F-statistic) 0.000014

  • 7/29/2019 Lecture 14 econometrics

    19/35

    Dependent Variable: LNWAGEMethod: Least SquaresDate: 11/01/01 Time: 08:47

    Sample: 1 49Included observations: 49

    Variable Coefficient Std. Error t-Statistic Prob.

    C 7.401588 0.160656 46.07096 0.0000GENDER 0.276639 0.074299 3.723311 0.0006EXPER 0.017241 0.004698 3.669938 0.0007EDUC 0.022429 0.013386 1.675632 0.1018

    AGE -0.002105 0.002676 -0.786674 0.4362RACE 0.095861 0.060972 1.572208 0.1240D23 -0.293381 0.187884 -1.561503 0.1265

    CLERICAL -0.419411 0.083845 -5.002245 0.0000CRAFTS -0.342397 0.081728 -4.189469 0.0002MAINT -0.525459 0.092669 -5.670253 0.0000

    R-squared 0.778514 Mean dependent var 7.454952Adjusted R-squared 0.727402 S.D. dependent var 0.312741S.E. of regression 0.163285 Akaike info criterion -0.606736Sum squared resid 1.039816 Schwarz criterion -0.220650Log likelihood 24.86503 F-statistic 15.23148Durbin-Watson stat 1.985725 Prob(F-statistic) 0.000000

  • 7/29/2019 Lecture 14 econometrics

    20/35

    Application: Election 2000 in Florida

    Effect of butterfly ballot in Palm Beach County on Buchanan vote

    Data for all Florida counties

    Votes candidates

    Size and demographic composition of counties (census). What isrelevant?

  • 7/29/2019 Lecture 14 econometrics

    21/35

    Model

    Dependent variable?

    Independent variables?

    How do we check whether Palm Beach is different?

  • 7/29/2019 Lecture 14 econometrics

    22/35

    Election 2000 in Florida: Butterfly ballot in Palm Beach county

    Outcome of 2000 presidential election disputed.

    Claims of voting irregularities in Florida.

    One issue was a confusing ballot design in Palm Beach county, thebutterfly ballot.

    Order of punch holes different from order of the two main candidates,Bush and Gore.

    Claim: Many voters mistakenly voted for Buchanan, the candidate of

    the Reform Party.

  • 7/29/2019 Lecture 14 econometrics

    23/35

    Research question: Did Buchanan get an unusually large fraction of thevotes in Palm Beach county?

  • 7/29/2019 Lecture 14 econometrics

    24/35

    Regression model

    Dependent variable: log of fraction votes for Buchanan.

    Independent variables

    Percentage of population Hispanic

    Percentage of population Black

    Percentage of population over 65

    Percentage of population with college degree

    Income (1000$ per year)

    Population (10000)

  • 7/29/2019 Lecture 14 econometrics

    25/35

    Descriptive statistics

    Date:04/06/05Sample: 1 67

    Time: 22:14

    FRACBUCHA FRACGORE FRACBUSH PERCBLACK PERCHISPAN PERCOVER6 PERCCOLLE INCOME1000 POPULATION

    Mean 0.004697 0.428125 0.551544 15.89701 6.288060 16.80293 13.89701 26.18864 21.87156Median 0.003976 0.430705 0.549881 14.40000 3.500000 14.60939 11.90000 25.71800 8.191900Maximum 0.017452 0.676075 0.741084 61.80000 54.40000 33.43856 37.10000 38.13000 204.4600Minimum 0.000897 0.241105 0.310129 2.300000 0.900000 6.974674 5.200000 17.09800 0.628900Std. Dev. 0.003218 0.091383 0.092058 11.07191 8.186436 7.011421 6.588534 4.794646 36.05383Skewness 1.912928 0.348309 -0.282254 1.926733 3.699223 0.846034 1.203425 0.446578 3.023152

    Kurtosis 7.448237 3.331569 3.187830 7.627821 19.94750 2.718083 4.690722 2.364601 13.30769

    J arque-Bera 96.10031 1.661643 0.988110 101.2424 954.6232 8.214687 24.15201 3.354070 398.6674Probability 0.000000 0.435691 0.610147 0.000000 0.000000 0.016451 0.000006 0.186927 0.000000

    Observations 67 67 67 67 67 67 67 67 67

  • 7/29/2019 Lecture 14 econometrics

    26/35

    OLS results: Basis equation

  • 7/29/2019 Lecture 14 econometrics

    27/35

    Signs of coefficients plausible?

    Interpretation of coefficients: dependent variable is log!

  • 7/29/2019 Lecture 14 econometrics

    28/35

    OLS residuals: Graph

    -1.0

    -0.5

    0.0

    0.5

    1.0

    1.5

    2.0

    5 10 15 20 25 30 35 40 45 50 55 60 65

    LNFRACBUCHANAN Residuals

  • 7/29/2019 Lecture 14 econometrics

    29/35

    OLS residuals: Table

  • 7/29/2019 Lecture 14 econometrics

    30/35

  • 7/29/2019 Lecture 14 econometrics

    31/35

    OLS results: Palm Beach dummy

    To check whether Palm Beach is special include dummy that is 1 for PalmBeach (observation 50) and 0 otherwise

  • 7/29/2019 Lecture 14 econometrics

    32/35

    Interpretation of Palm Beach dummy

    dy 794.1521.2ln ++= "

    Hence

    794.1lnln = normalobserved yy

    so that

    103.51

    794.1

    ==

    ey

    yy

    normal

    normalobserved

    i.e. fraction 5 times higher than expected. Fraction is .00789.

  • 7/29/2019 Lecture 14 econometrics

    33/35

    Sensitivity check: Include log fraction Bush vote

  • 7/29/2019 Lecture 14 econometrics

    34/35

    Effect on Bush and Gore vote

  • 7/29/2019 Lecture 14 econometrics

    35/35