Yining Chen

Smoothed log-concave maximum likelihood estimation with applications

We study the smoothed log-concave maximum likelihood estimator of a probability distribution on R d . This is a fully automatic nonparametric density estimator, obtained as a canonical smoothing of the log-concave maximum likelihood estimator. We demonstrate its attractive features both through an analysis of its theoretical properties and a simulation study. Moreover, we show how the estimator can be used as an intermediate stage of more involved procedures, such as constructing a classifier or estimating a functional of the density. Here again, the use of the estimator can be justified both on theoretical grounds and through its finite sample performance, and we illustrate its use in a breast cancer diagnosis (classification) problem.

Alloantibody responses after renal transplant failure can be better predicted by donor-recipient HLA amino acid sequence and physicochemical disparities than conventional HLA matching

We have assessed whether HLA immunogenicity as defined by differences in donor–recipient HLA amino‐acid sequence (amino‐acid mismatch score, AMS; and eplet mismatch score, EpMS) and physicochemical properties (electrostatic mismatch score, EMS) enables prediction of allosensitization to HLA, and also prediction of the risk of an individual donor–recipient HLA mismatch to induce donor‐specific antibody (DSA). HLA antibody screening was undertaken using single‐antigen beads in 131 kidney transplant recipients returning to the transplant waiting list following first graft failure. The effect of AMS, EpMS, and EMS on the development of allosensitization (calculated reaction frequency [cRF]) and DSA was determined. Multivariate analyses, adjusting for time on the waiting list, maintenance on immunosuppression after transplant failure, and graft nephrectomy, showed that AMS (odds ratio [OR]: 1.44 per 10 units, 95% CI: 1.02–2.10, p = 0.04) and EMS (OR: 1.27 per 10 units, 95% CI: 1.02–1.62, p = 0.04) were independently associated with the risk of developing sensitization to HLA (cRF > 15%). AMS, EpMS, and EMS were independently associated with the development of HLA‐DR and HLA‐DQ DSA, but only EMS correlated with the risk of HLA‐A and ‐B DSA development. Differences in donor–recipient HLA amino‐acid sequence and physicochemical properties enable better assessment of the risk of HLA‐specific sensitization than conventional HLA matching.

Early Graft Loss After Kidney Transplantation: Risk Factors and Consequences

Early graft loss (EGL) after kidney transplantation is a catastrophic outcome that is assumed to be more likely after the use of kidneys from suboptimal donors. We therefore examined its incidence, risk factors and consequences in our center in relation to different donor types. Of 801 recipients who received a kidney‐only transplant from deceased donors, 50 (6.2%) suffered EGL within 30 days of transplantation. Significant risks factors for EGL were donation after circulatory death (DCD) (odds ratio [OR] 2.88; p = 0.006), expanded criteria donor (ECD) transplantation (OR 4.22; p = 0.010), donor age (OR 1.03; p = 0.044) and recipient past history of thrombosis (OR 4.91; p = 0.001). Recipients with EGL had 12.28 times increased risk of death within the first year, but long‐term survival was worse for patients remaining on the waiting list. In comparison with patients on the waiting list but not transplanted, and with all patients on the waiting list, the risk of death after EGL decreased to baseline 4 and 23 months after transplantation, respectively. Our findings suggest that DCD and ECD transplantation are significant risk factors for EGL, which is a major risk factor for recipient death. However, long‐term mortality is even greater for those remaining on the waiting list.

On convex least squares estimation when the truth is linear

We prove that the convex least squares estimator (LSE) attains a n-1/2 pointwise rate of convergence in any region where the truth is linear. In addition, the asymptotic distribution can be characterized by a modified invelope process. Analogous results hold when one uses the derivative of the convex LSE to perform derivative estimation. These asymptotic results facilitate a new consistent testing procedure on the linearity against a convex alternative. Moreover, we show that the convex LSE adapts to the optimal rate at the boundary points of the region where the truth is linear, up to a log-log factor. These conclusions are valid in the context of both density estimation and regression function estimation.

Shape constrained kernel-weighted least squares: Estimating production functions for Chilean manufacturing industries

ABSTRACT In this article, we examine a novel way of imposing shape constraints on a local polynomial kernel estimator. The proposed approach is referred to as shape constrained kernel-weighted least squares (SCKLS). We prove uniform consistency of the SCKLS estimator with monotonicity and convexity/concavity constraints and establish its convergence rate. In addition, we propose a test to validate whether shape constraints are correctly specified. The competitiveness of SCKLS is shown in a comprehensive simulation study. Finally, we analyze Chilean manufacturing data using the SCKLS estimator and quantify production in the plastics and wood industries. The results show that exporting firms have significantly higher productivity.