st: RE: Allowing different lag lengths by country in an error correction model with panel data

["Millimet, Daniel" <millimet@mail.smu.edu>]

There is a lengthy literature in time series models with irregular spacing.  The literature on dynamic panel models is smaller and the problem is complex.

Dufour, J.-M. and M.G. Dagenais (1985), "Durbin-Watson Tests for Serial Correlation in Regressions with Missing Observations," Journal of Econometrics, 27, 371-381.
Dunsmuir, W. and P.M. Robinson (1981), "Estimation of Time Series Models in the Presence of Missing Data," Journal of the American Statistical Association, 76, 560-568.
Harvey, A.C. and R.G. Pierse (1984), "Estimating Missing Observations in Economic Time Series," Journal of the American Statistical Association, 79, 125-131
Jones, R.H. (1980), "Maximum Likelihood Fitting of ARMA Models to Time Series with Missing Observations," Technometrics, 22, 389-395.
Jones, R.H. (1985), "Time Series Analysis with Unequally Spaced Data," in E.J. Hannan, P.R. Krishnaiah, and M.M. Rao (eds) Handbook of Statistics, Vol. 5: Time Series in the Time Domain, New York: North-Holland, 157-177.
Jones, R.H. (1986), "Time Series Regression with Unequally Spaced Data," Journal of Applied Probability, 23, 89-98.
Jones, R.H. and F. Boadi-Boateng (1991), "Unequally Spaced Longitudinal Data with AR(1) Serial Correlation," Biometrics, 47, 161-175.
Kohn, R. and C.F. Ansley (1986), "Estimation, Prediction, and Interpolation for ARIMA Models with Missing Data," Journal of the American Statistical Association, 81, 751-761.
Palm, F.C. and T.E. Nijman (1984), "Missing Observations in the Dynamic Regression Model," Econometrica, 52, 1415-1435.
Robinson, P.M. (1985), "Testing for Serial Correlation in Regression with Missing Observations," Journal of the Royal Statistical Society, Series B, 47, 429-437.
Rosner, B. and A. Munoz (1988), "Autoregressive Modelling for the Analysis of Longitudinal Data with Unequally Spaced Examinations," Statistics in Medicine, 7, 59-71.
Ryan, K.F. and D.E.A. Giles (1998), "Testing for Unit Roots in Economic Time-Series with Missing observations," in T. B. Fomby and R. C. Hill (eds.) Advances in Econometrics, 13, 203-242.
Savin, N.E. and K.T. White (1978), "Testing for Autocorrelation with Missing Observations," Econometrica, 46, 59-67.
Shin, D.W. and S. Sarkar (1994a), "Unit Roots for ARIMA(0,1,q) Models with Irregularly Observed Samples," Statistics and Pobability Letters, 19, 188-194.
Shin, D.W. and S. Sarkar (1994b), "Likelihood Ratio Type Unit Root Tests for AR(1) Models with Nonconsecutive Observations," Comunications in Statistics: Theory and Methods, 23, 1387-1397
Shively, T.S. (1993), "Testing for Autoregressive Disturbances in a Time Series Regression with Missing Observations," Journal of Econometrics, 57, 233-255.

****************************************************
Daniel L. Millimet, Professor
Department of Economics
Box 0496
SMU
Dallas, TX 75275-0496
phone: 214.768.3269
fax: 214.768.1821
web: http://faculty.smu.edu/millimet
****************************************************

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Tejaswi Velayudhan (tvelayudhan@cgdev.org)
Sent: Tuesday, February 12, 2013 2:55 PM
To: statalist@hsphsun2.harvard.edu
Subject: st: Allowing different lag lengths by country in an error correction model with panel data

I apologize, reposting because of a formatting error: 

I have a country level dataset with data from the years in which a demographic health survey (DHS) was run in each country. This means it's sort of an unbalanced panel dataset, with the catch that the length of the gap between years of the series is different for each country. I would like to use this data to run an error-correction or auto-regressive distributed lag model but instead of controlling for y_t-1 and x_t-1 on the right hand side, I would be adjusting for  y_t-k and x_t-k, where k varies by country. I have been looking for a way to adjust for the length of the gap (k), but surprisingly cannot find any literature on this or even papers that have dealt with this sort of data. Perhaps  I am just not looking for the right keywords but I thought that someone on this listserv may be able to point me in the right direction. 

One possibility I have considered is imputing the t-1 value of the variable based on the t-k value with some discounting factor assumed as a function of k. Is there a more straightforward adjustment of the ECM or ARDL model that allows you to work with t-k instead of t-1? 

Thanks,

Teju

Tejaswi Velayudhan
Research Assistant
Center for Global Development


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/