Comparing the efficiency of control charts when using three different estimators for fraction of nonconforming 

Supervisor: Dr Julio Pereira

Supervisor’s webpage: Dr Julio Pereira – Senior Tutor – Statistics – Massey University

In the literature there are at least three known estimators for the proportion p of nonconforming items in a sample when the sample size is variable. A comparative study has not been found in the literature with regards the choice of the best estimator to be used in the calculation of the control limits for control charts of fraction of nonconforming.  In other words, the effect of using one or another estimator in detecting an out-of-control process is unknown. The aim of this project is to carry out a simulation study to compare the impact of using three ratio estimators for nonconforming fraction in the calculation of the control chart limits when the sample size is not fixed. In this way we will be able to find out if there is any advantage or disadvantage in using any of the three estimators in various scenarios. 

Please feel free to contact Dr Julio Pereira by email: J.Pereira@massey.ac.nz if you have any questions about this project.

Analyzing the impact of COVID on scientific publishing

Supervisor: Professor Thomas Pfeiffer

Supervisor’s webpage: Prof Thomas Pfeiffer – Professor in Computational Biology/Biochemistry – Massey University

The COVID pandemic had a substantial impact in how scientific results are disseminated. Instead of being published after comprehensive peer review, many scientific findings on COVID are published on so-called preprint servers before they are peer reviewed. This accelerates the speed at which relevant findings become accessible to peers and decision makers, but also poses the risk as preprints might be less reliable than peer-reviewed publications. We are working with a sample of 400 COVID preprints to understand impact of COVID on scientific publishing. This project is relevant for undergrads with interest in data science, statistics, and/or biomedical research. The timeline is flexible.