Unweighted estimation based on optimal sample under measurement constraints

TL;DR

This paper introduces an unweighted estimation method based on optimal subsampling under measurement constraints, improving efficiency in parameter estimation for large datasets with costly responses.

Contribution

It proposes a novel unweighted estimator using optimal subsamples and derives its asymptotic distribution without relying on a pilot estimate, enhancing estimation efficiency.

Findings

Unweighted estimator outperforms weighted methods in efficiency.

Asymptotic distribution derived using martingale techniques.

Numerical results confirm improved performance.

Abstract

To tackle massive data, subsampling is a practical approach to select the more informative data points. However, when responses are expensive to measure, developing efficient subsampling schemes is challenging, and an optimal sampling approach under measurement constraints was developed to meet this challenge. This method uses the inverses of optimal sampling probabilities to reweight the objective function, which assigns smaller weights to the more important data points. Thus the estimation efficiency of the resulting estimator can be improved. In this paper, we propose an unweighted estimating procedure based on optimal subsamples to obtain a more efficient estimator. We obtain the unconditional asymptotic distribution of the estimator via martingale techniques without conditioning on the pilot estimate, which has been less investigated in the existing subsampling literature. Both asymptotic results and numerical results show that the unweighted estimator is more efficient in parameter estimation.