1 / 21

VAR and VEC

VAR and VEC. Using Stata. VAR: Vector Autoregression. Assumptions: y t : Stationary K-variable vector v : K constant parameters vector A j : K by K parameters matrix, j=1,…,p u t : i.i.d.( 0 , S ) Exogenous variables X may be added. VAR and VEC.

issac
Download Presentation

VAR and VEC

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. VAR and VEC Using Stata

  2. VAR: Vector Autoregression • Assumptions: • yt: Stationary K-variable vector • v: K constant parameters vector • Aj: K by K parameters matrix, j=1,…,p • ut: i.i.d.(0,S) • Exogenous variables X may be added

  3. VAR and VEC • If yt is not stationary, VAR or VEC can only be applied for cointegrated yt system: • VAR (Vector Autoregression) • VEC (Vector Error Correction)

  4. VEC: Vector Error Correction • If there is no trend in yt, let P = ab’ (P is K by K, a is K by r, b is K by r, r is the rank of P, 0<r<K):

  5. VEC: Vector Error Correction • No-constant or No-drift Model: g = 0, m = 0 (or v = 0) • Restricted-constant Model: g = 0, m≠ 0 (or v = am) • Constant or Drift Model: g≠ 0, m≠ 0

  6. VEC: Vector Error Correction • If there is trend in yt

  7. VEC: Vector Error Correction • No-drift No-trend Model: g = 0, m = 0, t = 0, r = 0 (or v = 0, d = 0) • Restricted-constant Model: g = 0, m≠ 0, t = 0, r = 0 (or v = am, d = 0) • Constant or Drift Model: g≠ 0, m≠ 0, t = 0, r = 0 (or d = 0) • Restricted-trend Model:g≠ 0, m≠ 0, t = 0, r≠ 0 (or d = ar) • Trend Model:g≠ 0, m≠ 0, t≠ 0, r≠ 0

  8. Example • C: Personal Consumption Expenditure • Y: Disposable Personal Income • C ~ I(1), Y ~ I(1) • Consumption-Income Relationship:Ct = a + bYt-1 + gCt-1 +…(+ dt) + e • Cointegrated C and I: e ~ I(0) • Johansen Test with constant model and 10 lags

  9. Example • VAR: • VEC: • Rank 1 P = ab’

  10. Example • Empirical Model: 1949q4 – 2006q4 • Constant, 10 lags • Restricted-constant, 10 lags

  11. Example (Y = PCE)

  12. Example (X = DPI)

  13. Unrestricted VAR

  14. Restricted VAR • Parameters Restrictions • Many of the parameters associated with augmented lags are 0 • Structural VAR • Linking with macroeconomic theory

  15. VAR and VMA

  16. Impulse Response Function • Based on VMA representation, we can trace out the time path of the various shocks on the variables in the VAR system.

  17. VAR and Simultaneous Equations • Identification: • A = -B-1G, ut = B-1et • A = [A1,…,Ap,v], G = [G1,…, Gp,g0] • B and G1,…,Gp are K by K parameters matrices • Var(ut) = S = B-1Var(et)B-1’

  18. Structural VAR • VAR as a reduced form of simultaneous equations model requires parameters restrictions so that the system model is identified or estimatable. • A structural VAR corresponds to an identified simultaneous equations system. • The restrictions are necessarily placed on the parameters matrix B and therefore the variance-covariance matrix S of VAR.

  19. Example • 2-Equation System Model: • Structural VAR:

  20. Example • Structural VAR: • Identification and Parameters Restrictions

  21. Stata Programs • us_dpi_pce.txt • dpi_pce4.do • dpi_pce5.do • dpi_pce6.do • dpi_pce7.do

More Related