English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Simulation is a crucial tool for the evaluation and comparison of statisticalmethods. How to design fair and neutral simulation studies is therefore ofgreat interest for both researchers developing new methods and practitionersconfronted with the choice of the most suitable method. The term simulationusually refers to parametric simulation, that is, computer experiments usingartificial data made up of pseudo-random numbers. Plasmode simulation, that is,computer experiments using the combination of resampling feature data from areal-life dataset and generating the target variable with a user-selectedoutcome-generating model (OGM), is an alternative that is often claimed toproduce more realistic data. We compare parametric and Plasmode simulation forthe example of estimating the mean squared error of the least squares estimatorin linear regression. If the true underlying data-generating process (DGP) andthe OGM were known, parametric simulation would be the best choice in terms ofestimating the MSE well. However, in reality, both are usually unknown, soresearchers have to make assumptions: in Plasmode simulation studies for theOGM, in parametric simulation for both DGP and OGM. Most likely, theseassumptions do not reflect the truth. Here, we aim to find out how assumptionsdeviating from the true DGP and the true OGM affect the performance ofparametric simulation and Plasmode simulations in the context of MSE estimationfor the least squares estimator and in which situations which simulation typeis preferable. Our results suggest that the preferable simulation methoddepends on many factors, including the number of features, and how theassumptions of a parametric simulation differ from the true DGP. Also, theresampling strategy used for Plasmode influences the results. In particular,subsampling with a small sampling proportion can be recommended.

When is Plasmode simulation superior to parametric simulation when estimating the MSE of the least squares estimator in linear regression?