More Like this

1. More Efficient Estimation for Logistic Regression with Optimal Subsamples

   Wang, HaiYing (August 2019, Journal of machine learning research)

   In this paper, we propose improved estimation method for logistic regression based on subsamples taken according the optimal subsampling probabilities developed in Wang et al. (2018). Both asymptotic results and numerical results show that the new estimator has a higher estimation efficiency. We also develop a new algorithm based on Poisson subsampling, which does not require to approximate the optimal subsampling probabilities all at once. This is computationally advantageous when available random-access memory is not enough to hold the full data. Interestingly, asymptotic distributions also show that Poisson subsampling produces a more efficient estimator if the sampling ratio, the ratio of the subsample size to the full data sample size, does not converge to zero. We also obtain the unconditional asymptotic distribution for the estimator based on Poisson subsampling. Pilot estimators are required to calculate subsampling probabilities and to correct biases in un-weighted estimators; interestingly, even if pilot estimators are inconsistent, the proposed method still produce consistent and asymptotically normal estimators. 

   more » « less
2. Determination of the asymptotic limits of adaptive photon counting measurements for coherent-state optical phase estimation

   https://doi.org/10.1038/s41534-022-00601-8  

   Rodríguez-García, M. A.; DiMario, M. T.; Barberis-Blostein, P.; Becerra, F. E. (December 2022, npj Quantum Information)

   Abstract Physical realizations of the canonical phase measurement for the optical phase are unknown. Single-shot phase estimation, which aims to determine the phase of an optical field in a single shot, is critical in quantum information processing and metrology. Here we present a family of strategies for single-shot phase estimation of coherent states based on adaptive non-Gaussian, photon counting, measurements with coherent displacements that maximize information gain as the measurement progresses, which have higher sensitivities over the best known adaptive Gaussian strategies. To gain understanding about their fundamental characteristics and demonstrate their superior performance, we develop a comprehensive statistical analysis based on Bayesian optimal design of experiments, which provides a natural description of these non-Gaussian strategies. This mathematical framework, together with numerical analysis and Monte Carlo methods, allows us to determine the asymptotic limits in sensitivity of strategies based on photon counting designed to maximize information gain, which up to now had been a challenging problem. Moreover, we show that these non-Gaussian phase estimation strategies have the same functional form as the canonical phase measurement in the asymptotic limit differing only by a scaling factor, thus providing the highest sensitivity among physically-realizable measurements for single-shot phase estimation of coherent states known to date. This work shines light into the potential of optimized non-Gaussian measurements based on photon counting for optical quantum metrology and phase estimation. 

   more » « less
3. Optimal subsampling for quantile regression in big data

   https://doi.org/10.1093/biomet/asaa043  

   Wang, Haiying; Ma, Yanyuan (July 2020, Biometrika)

   null (Ed.)

   Summary We investigate optimal subsampling for quantile regression. We derive the asymptotic distribution of a general subsampling estimator and then derive two versions of optimal subsampling probabilities. One version minimizes the trace of the asymptotic variance-covariance matrix for a linearly transformed parameter estimator and the other minimizes that of the original parameter estimator. The former does not depend on the densities of the responses given covariates and is easy to implement. Algorithms based on optimal subsampling probabilities are proposed and asymptotic distributions, and the asymptotic optimality of the resulting estimators are established. Furthermore, we propose an iterative subsampling procedure based on the optimal subsampling probabilities in the linearly transformed parameter estimation which has great scalability to utilize available computational resources. In addition, this procedure yields standard errors for parameter estimators without estimating the densities of the responses given the covariates. We provide numerical examples based on both simulated and real data to illustrate the proposed method. 

   more » « less
4. Asymptotic analysis of a new low complexity encryption approach for the Internet of Things, smart cities and smart grid

   https://doi.org/10.1109/ICSGSC.2017.8038576  

   Samudrala, Ananth Narayan; Blum, Rick S. (July 2017, 2017 IEEE International Conference on Smart Grid and Smart Cities (ICSGSC))

   Abstract (WSN) using encrypted non-binary quantized data is studied. In a WSN, sensors transmit their observations to a fusion center through a wireless medium where the observations are susceptible to unauthorized eavesdropping. Encryption approaches for WSNs with fixed threshold binary quantization were previously explored. However, fixed threshold binary quantization limits parameter estimation to scalar parameters. In this paper, we propose a stochastic encryption approach for WSNs that can operate on non-binary quantized observations and has the capability for vector parameter estimation. We extend a binary stochastic encryption approach proposed previously, to a nonbinary generalized case. Sensor outputs are quantized using a quantizer with R + 1 levels, where R in {1.2. 3 ...}, encrypted by flipping them with certain flipping probabilities, and then transmitted. Optimal estimators using maximum-likelihood estimation are derived for both a legitimate fusion center (LFC) and a third party fusion center (TPFC) perspectives. We assume the TPFC is unaware of the encryption. Asymptotic analysis of the estimators is performed by deriving the Cramer-Rao lower bound for LFC estimation, and the asymptotic bias and variance for TPFC estimation. Numerical results validating the asymptotic analysis are presented. 

   more » « less
5. On propensity score matching with a diverging number of matches

   https://doi.org/10.1093/biomet/asae026  

   He, Yihui; Han, Fang (June 2024, Biometrika)

   Abstract This paper re-examines the work of Abadie & Imbens (2016) on propensity score matching for average treatment effect estimation. We explore the asymptotic behaviour of these estimators when the number of nearest neighbours, M, grows with the sample size. It is shown, while not surprising, but technically nontrivial, that the modified estimators can improve upon the original fixed M-estimators in terms of efficiency. Additionally, we demonstrate the potential to attain the semiparametric efficiency lower bound when the propensity score admits some special structures, echoing the insight of Hahn (1998). 

   more » « less