Nonparametric estimation of a biometric function using neoteric ranked set sampling with application to breast cancer...

The mean residual life (MRL) function is one of the fundamental concepts in survival analysis. The MRL is easily interpretable and particularly useful for conveying the survival information to patients. Ranked set sampling (RSS) is a sampling design useful for situations where precise quantification of units is costly or difficult, but ranking them based on associated variables is easy with negligible cost. Health studies can be expensive and time-consuming due to the need for costly equipment or long follow-up periods. In such cases, RSS can serve as a suitable alternative to the commonly used sampling method, simple random sampling (SRS). In recent years, different modifications of RSS have been introduced to enhance its efficiency. This paper investigates the estimation of the MRL function using Neoteric Ranked Set Sampling (NRSS), a notable modification of RSS. We compare the proposed estimator with its counterparts in RSS and SRS using Monte Carlo simulation. Our results show that the proposed estimator is more efficient than its counterparts for small and moderate sample sizes. Additionally, we apply the proposed procedure to a real dataset of survival time of patients from the Surveillance, Epidemiology, and End Results (SEER) program of the National Cancer Institute (NCI) to estimate the MRL function for patients diagnosed with malignant breast cancer. Our findings show that using the NRSS technique improves the estimation process, and therefore a smaller sample size is needed to achieve a predetermined level of precision.