Volume : III, Issue : VI, June - 2013

Distributed Lags: A New Estimator

Prof. V. Balakrishnama Naidu, Dr. P. Surya Kumar, Dr . M. Sateeshnadha Reddy

Abstract :

‘Lagged variables’ is one of the basic problems in Econometrics. Under Standard assumptions of a general linear model, straight forward application of Ordinary Least Squares (OLS) procedure to obtain estimates of the parameters of the distributed lag model has two main disadvantages, they are (i) If the number of lags is large and the sample size is small, then the parameters may not be estimable; (ii) Since the successive values of the same explanatory variables are highly correlated, the problem of multicollinearity creeps in. Therefore, certain weight patterns are suggested to reduce the number of lagged variables and to alleviate the severity of multicollinearity (Lankipalle, 1977). Though there is good number of weight generating mechanisms in the literature, often Almon’s polynomial type weight pattern is used. Some Monte Carlo studies revealed that the OLS is supposed to be better than Almon’s procedure from the point of view of bias, and variance. In view of the above, in this paper an attempt has been made for estimation of parameters in Lagged variables linear models. A new estimator has been suggested and it is shown that the new estimator has smaller variance than Almon’s estimator and provides a remedy for multicollinearity.  

Keywords :

Article: Download PDF   DOI : 10.36106/ijar  

Cite This Article:

Prof. V. Balakrishnama Naidu, Dr. P. Surya Kumar, Dr .M. Sateeshnadha Reddy Distributed Lags: A New Estimator Indian Journal of Applied Research, Vol.III, Issue.VI June 2013

Number of Downloads : 540

References :