-------------------------------------------------------------------------------
help: lmadw                                                        dialog: lmad
> w
-------------------------------------------------------------------------------

+-------+ ----+ Title +------------------------------------------------------------

lmadw: Durbin-Watson Autocorrelation Test at Higher Order AR(p)

+--------+ ----+ Syntax +-----------------------------------------------------------

lmadw depvar indepvars [if] [in] [weight] , [ lags(numlist) noconstant vce(vcetype) ]

+---------+ ----+ Options +----------------------------------------------------------

lags(#) determine Order of Lag Length; default is lag(1).

noconstant suppress constant term

SE/Robust vce(vcetype) vcetype may be ols, robust, cluster clustvar, bootstrap, jackknife, hc2, or hc3

+-------------+ ----+ Description +------------------------------------------------------

lmadw computes Durbin-Watson Autocorrelation Test after regress command. lmadw detects autocorrelation at Higher Order AR(p), more than AR(1).

{cmd: DW(i): Durbin-Watson Test = sum((E-`E'[n-i])^2)/sum(E^)

************************************************************** ***** Positive Auto ***** (DW Test) **** Negative Auto ******* * DW > Du (No+) * DW * 4-Du (No-) * * 0 < DW < DL ( +) * 4-DL < DW < 4 ( -) * * DL * DW * Du (Inc) * 4-Du <= DW * 4-DL (Inc) * **************************************************************

+---------------+ ----+ Saved Results +----------------------------------------------------

lmadurh saves the following in r():

Scalars r(rho_#) Rho Value at Order AR(i) r(dw_#) Durbin-Watson Test at Order AR(i)

+----------+ ----+ Examples +---------------------------------------------------------

clear all

db lmadw

sysuse lmadw.dta , clear

lmadw y x1 x2 , lags(1)

lmadw y x1 x2 , lags(4)

return list

================================================ * Durbin-Watson Autocorrelation Test * ================================================ Ho: No Autocorrelation - Ha: Autocorrelation

--------------------------------------------------------------------------- * Rho Value for AR(1) = -0.1455 * Durbin-Watson Test AR(1) = 2.0185 df: (3 , 17) --------------------------------------------------------------------------- * Rho Value for AR(2) = -0.2231 * Durbin-Watson Test AR(2) = 2.0359 df: (3 , 17) --------------------------------------------------------------------------- * Rho Value for AR(3) = 0.1871 * Durbin-Watson Test AR(3) = 1.1956 df: (3 , 17) --------------------------------------------------------------------------- * Rho Value for AR(4) = -0.3002 * Durbin-Watson Test AR(4) = 2.0133 df: (3 , 17) ---------------------------------------------------------------------------

+------------+ ----+ References +-------------------------------------------------------

Durbin, James & Watson G (1950) "Testing for Serial Correlation in Least Square Regression", Biometrika, vol.37; 409-428.

Durbin, James & Watson G (1951) "Testing for Serial Correlation in Least Square Regression", Biometrika, Vol. 38; 159-178.

Maddala, G. (1992) "Introduction to Econometrics", 2nd ed., Macmillan Publishing Company, New York, USA; 245.

+--------+ ----+ Author +-----------------------------------------------------------

Emad Abd Elmessih Shehata Assistant Professor Agricultural Research Center - Agricultural Economics Research Institute - Eg > ypt Email: emadstat@hotmail.com WebPage: http://emadstat.110mb.com/stata.htm WebPage at IDEAS: http://ideas.repec.org/f/psh494.html WebPage at EconPapers: http://econpapers.repec.org/RAS/psh494.htm

+----------------+ ----+ lmadw Citation +---------------------------------------------------

Shehata, Emad Abd Elmessih (2011) "lmadw: Stata Module to Compute Durbin-Watson Autocorrelation Test at Higher Order AR(p) after OLS Regression"

Also see

Online: lmareg3, lmadurh, lmalb, lmabp, lmadw, lmavon (if installed).