Enflasyon forecast

  • Published on
    21-Feb-2017

  • View
    59

  • Download
    0

Embed Size (px)

Transcript

  • TKETC FYAT ENDEKS

    TAHMN MODEL

  • SUNUMZaman Serilerinde Duraanlk

    Analiz Kapsam

    Zaman Serilerinin Dzey Halleri

    Zaman Serilerinin Korelogram Analizi

    Logaritmik Dnm Yaplm Zaman Serileri

    Logaritmik Dnm Yaplm Zaman Serilerinin Birinci Derecen Farklar

    Fark Serileriyle Kurulan Dorusal Modelin En Kk Kareler Yntemiyle Regresyon Analizi

    Breusch Godfrey Otokorelayon Testi

    Deien Varyans Analizi in White Testi

    Yapsal Krlma Testleri (Cusum / Cusum of Squares)

    Jarque-Bera Normal Dalm Testi

    Engle ve Yoo (1987) Etmleme Analizi

    Statik (Ex-post) ngr

    Dinamik (Ex-ante) ngr

  • Zaman Serilerinde Duraanlk

    = 0

    = 2

    , = 0

    ~

    ()

  • Analiz Kapsam

    Deikenler

    : Tketici Fiyat Endeksi (TFE) (2003=100): Tketici Gven Endeksi: 2 Yl Vadeli Devlet Tahvili: M2 Para Arz

    Gzlem Says : 122Frekans : AylkZaman Aral : Aralk 2005 Ocak 2016Kullanlan Yntem : En Kk Kareler Yntemi (EKK)

  • Zaman Serilerinin Dzey Halleri

  • Zaman Serilerinin Korelogram Analizi

    Tfe

  • Zaman Serilerinin Korelogram Analizi

    M2 Para Arz

  • Zaman Serilerinin Korelogram Analizi

    Gven Endeksi

  • Zaman Serilerinin Korelogram Analizi

    2 Yl Vadeli Devlet Tahvili

  • Logaritmik Dnm Yaplm Zaman Serileri

  • Logaritmik Dnm Yaplm Zaman Serilerinin Birinci Derecen Farklar

  • Model 1: Fark Serileriyle Kurulan Dorusal Modelin En Kk Kareler

    Yntemiyle Regresyon Analizi

    Dependent Variable: DLTUFE

    Method: Least Squares

    Date: 03/03/16 Time: 22:14

    Sample (adjusted): 2006M05 2016M01

    Included observations: 117 after adjustments

    Variable Coefficient Std. Error t-Statistic Prob.

    DLM2 -0.112835 0.046330 -2.435437 0.0165

    DLGUVEN -0.045788 0.017268 -2.651685 0.0092

    DLTAHVIL 0.001577 0.000622 2.534316 0.0126

    C 0.009981 0.001074 9.296794 0.0000

    DLTUFE(-4) -0.284613 0.086239 -3.300287 0.0013

    R-squared 0.199790 Mean dependent var 0.006664

    Adjusted R-squared 0.171211 S.D. dependent var 0.007988

    S.E. of regression 0.007272 Akaike info criterion -6.967812

    Sum squared resid 0.005923 Schwarz criterion -6.849770

    Log likelihood 412.6170 Hannan-Quinn criter. -6.919888

    F-statistic 6.990815 Durbin-Watson stat 1.806012

    Prob(F-statistic) 0.000046

    = 0,009981 0,1128352 0,045788 + 0,001577 0,2846134

    )0.001074 0.046330 0.017268 0.000622 (0.086239

    )9,29 2,43 2,65 2,53 (3,30

  • Model 1: Breusch Godfrey Otokorelayon Testi

    Breusch-Godfrey Serial Correlation LM Test:

    F-statistic 2.105190 Prob. F(2,110) 0.1267

    Obs*R-squared 4.313220 Prob. Chi-Square(2) 0.1157

    Test Equation:

    Dependent Variable: RESID

    Method: Least Squares

    Date: 03/16/16 Time: 23:41

    Sample: 2006M05 2016M01

    Included observations: 117

    Presample missing value lagged residuals set to zero.

    Variable Coefficient Std. Error t-Statistic Prob.

    DLM2 0.017220 0.046711 0.368661 0.7131

    DLGUVEN -0.002503 0.017240 -0.145166 0.8848

    DLTAHVIL -0.000264 0.000642 -0.411482 0.6815

    C -6.97E-05 0.001072 -0.064986 0.9483

    DLTUFE(-4) -0.025378 0.087639 -0.289571 0.7727

    RESID(-1) 0.099728 0.098900 1.008366 0.3155

    RESID(-2) -0.185257 0.098423 -1.882259 0.0624

    R-squared 0.036865 Mean dependent var -2.97E-20

    Adjusted R-squared -0.015670 S.D. dependent var 0.007145

    S.E. of regression 0.007201 Akaike info criterion -6.971186

    Sum squared resid 0.005704 Schwarz criterion -6.805927

    Log likelihood 414.8144 Hannan-Quinn criter. -6.904093

    F-statistic 0.701730 Durbin-Watson stat 1.957250

    Prob(F-statistic) 0.648789

    0: 1:

    0 . 1

  • Model 1: Deien Varyans Analizi in White Testi*

    0: 1 1: 1

    0 . 1

    Heteroskedasticity Test: White

    F-statistic 1.228216 Prob. F(14,102) 0.2669

    Obs*R-squared 16.87837 Prob. Chi-Square(14) 0.2627

    Scaled explained SS 12.54079 Prob. Chi-Square(14) 0.5630

    Test Equation:

    Dependent Variable: RESID^2

    Method: Least Squares

    Date: 03/16/16 Time: 23:42

    Sample: 2006M05 2016M01

    Included observations: 117

    Variable Coefficient Std. Error t-Statistic Prob.

    C 5.30E-05 1.17E-05 4.525277 0.0000

    DLM2^2 -0.004903 0.017520 -0.279825 0.7802

    DLM2*DLGUVEN 0.011040 0.013927 0.792703 0.4298

    DLM2*DLTAHVIL 0.000298 0.000401 0.742982 0.4592

    DLM2*DLTUFE(-4) 0.052604 0.060424 0.870583 0.3860

    DLM2 2.57E-05 0.000803 0.032030 0.9745

    DLGUVEN^2 0.000320 0.002087 0.153414 0.8784

    DLGUVEN*DLTAHVIL -0.000125 0.000143 -0.875187 0.3835

    DLGUVEN*DLTUFE(-4) 0.035278 0.025120 1.404354 0.1633

    DLGUVEN -0.000138 0.000283 -0.488810 0.6260

    DLTAHVIL^2 1.33E-06 2.82E-06 0.471393 0.6384

    DLTAHVIL*DLTUFE(-4) -0.000669 0.000946 -0.707281 0.4810

    DLTAHVIL 5.40E-06 1.12E-05 0.482438 0.6305

    DLTUFE(-4)^2 -0.052176 0.069912 -0.746301 0.4572

    DLTUFE(-4) -0.000591 0.001488 -0.397209 0.6920

    R-squared 0.144260 Mean dependent var 5.06E-05

    Adjusted R-squared 0.026805 S.D. dependent var 6.47E-05

    S.E. of regression 6.39E-05 Akaike info criterion -16.36037

    Sum squared resid 4.16E-07 Schwarz criterion -16.00624

    Log likelihood 972.0815 Hannan-Quinn criter. -16.21660

    F-statistic 1.228216 Durbin-Watson stat 1.635594

    Prob(F-statistic) 0.266914

    White testi, normal dalm koullu aramamaktadr.

  • Model 1: Yapsal Krlma Testleri (Cusum / Cusum of Squares)

    -40

    -30

    -20

    -10

    0

    10

    20

    30

    40

    2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

    CUSUM 5% Significance

    -0.2

    0.0

    0.2

    0.4

    0.6

    0.8

    1.0

    1.2

    2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

    CUSUM of Squares 5% Significance

  • Model 1: Jarque-Bera Normal Dalm Testi

    0

    2

    4

    6

    8

    10

    12

    -0.015 -0.010 -0.005 0.000 0.005 0.010 0.015

    Series: Residuals

    Sample 2006M05 2016M01

    Observations 117

    Mean -2.97e-20

    Median -0.000943

    Maximum 0.016451

    Minimum -0.018257

    Std. Dev. 0.007145

    Skewness 0.003546

    Kurtosis 2.621661

    Jarque-Bera 0.698055

    Probability 0.705374

    0: 1 .

    1: 1 .

    0 . 1 .( )

  • Model 2: Engle ve Yoo (1987) Etmleme Analizi

    0: = 0, . ~ 11: < 0, . ~ (0)

    = 1 + 22 + 3 + 4 + 54 +

    = 1 + =1

    + )~(0, 2

    )~(0, 2

    Dependent Variable: DKAL1

    Method: Least Squares

    Date: 03/03/16 Time: 22:35

    Sample (adjusted): 2006M11 2016M01

    Included observations: 111 after adjustments

    Variable Coefficient Std. Error t-Statistic Prob.

    KAL1(-1) -0.905691 0.093870 -9.648303 0.0000

    DKAL1(-5) -0.143060 0.067824 -2.109270 0.0372

    R-squared 0.466189 Mean dependent var 1.98E-05

    Adjusted R-squared 0.461292 S.D. dependent var 0.009301

    S.E. of regression 0.006827 Akaike info criterion -7.118082

    Sum squared resid 0.005080 Schwarz criterion -7.069261

    Log likelihood 397.0535 Hannan-Quinn criter. -7.098277

    Durbin-Watson stat 1.887883

    =

    = 9,64 < = 4,22

    0 . .2, 3, 4, 5 .

  • Statik (Ex-post) ngr

    4.8

    4.9

    5.0

    5.1

    5.2

    5.3

    5.4

    5.5

    5.6

    5.7

    2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

    LTUFE LTUFES

    , .

  • Dinamik (Ex-ante) ngr