I was saddened to learn that Anirudh Lal Nagar passed away on 4 February 2014.
Nagar was an exceptional Indian statistician and econometrician who made many fundamental contributions to our discipline. He was 83 years old at the time of his death.
Nagar (left) at the inauguration of the laboratory named in his honour at Jawaharlal Nehru University.
Nagar's work on the finite-sample inference in econometrics is especially well known. The term "the Nagar expansion" was coined by Denis Sargan (1974). This technique, proposed by Nagar in 1959 for the k-class estimators, was widely used to determine the sampling properties (bias, variance, etc.) of various simultaneous equations estimators. It came at a time when large-n asymptotics dominated the scene, and finite-sample results were deemed to be "intractable".
Nagar's work influenced a generation of theoretical econometricians, and paved the way for some of the most important results established in our discipline. The impact of his work is reflected in the volume of papers assembled to honour him on is sixtieth birthday (Carter et al., 1990).
He will be greatly missed.
References
Carter, R. A. L., J. Dutta, and A. Ullah (eds.), 1990. Contributions to Econometric Theory and Applications: Essays in Honour of A. L. Nagar. Wiley, New York. (Softcover reprint.)
Nagar, A. L., 1959. The bias and moment matrix of the general k-class estimators of the parameters in simultaneous equations. Econometrica, 27, 573-595.
Sargan, J. D., 1974. The validity of Nagar's expansion for the moments of econometric estimators. Econometrica, 42, 169-176.
