| _version_ | 1867004755872579584 |
|---|---|
| author | Alimu Dayimu Nikola Simidjievski Nikolaos Demiris Jean Abraham |
| author_facet | Alimu Dayimu Nikola Simidjievski Nikolaos Demiris Jean Abraham Alimu Dayimu Nikola Simidjievski Nikolaos Demiris Jean Abraham |
| collection | Wiley Open Access |
| contents | Sample size determination for prediction models via learning‐type curves Alimu Dayimu Nikola Simidjievski Nikolaos Demiris Jean Abraham Statistics in Medicine This article is concerned with sample size determination methodology for prediction models. We propose to combine the individual calculations via learning‐type curves. We suggest two distinct ways of doing so, a deterministic skeleton of a learning curve and a Gaussian process centered upon its deterministic counterpart. We employ several learning algorithms for modeling the primary endpoint and distinct measures for trial efficacy. We find that the performance may vary with the sample size, but borrowing information across sample size universally improves the performance of such calculations. The Gaussian process‐based learning curve appears more robust and statistically efficient, while computational efficiency is comparable. We suggest that anchoring against historical evidence when extrapolating sample sizes should be adopted when such data are available. The methods are illustrated on binary and survival endpoints. 10.1002/sim.10121 http://creativecommons.org/licenses/by/4.0/ |
| doi_str_mv | 10.1002/sim.10121 |
| format | Artículo Open Access |
| id | wiley_oa_10_1002_sim_10121 |
| institution | Wiley Open Access |
| license_str_mv | http://creativecommons.org/licenses/by/4.0/ |
| publishDate | 2024 |
| publisher | Wiley |
| record_format | wiley_oa |
| spellingShingle | Sample size determination for prediction models via learning‐type curves Alimu Dayimu Nikola Simidjievski Nikolaos Demiris Jean Abraham Statistics in Medicine Sample size determination for prediction models via learning‐type curves Alimu Dayimu Nikola Simidjievski Nikolaos Demiris Jean Abraham Statistics in Medicine This article is concerned with sample size determination methodology for prediction models. We propose to combine the individual calculations via learning‐type curves. We suggest two distinct ways of doing so, a deterministic skeleton of a learning curve and a Gaussian process centered upon its deterministic counterpart. We employ several learning algorithms for modeling the primary endpoint and distinct measures for trial efficacy. We find that the performance may vary with the sample size, but borrowing information across sample size universally improves the performance of such calculations. The Gaussian process‐based learning curve appears more robust and statistically efficient, while computational efficiency is comparable. We suggest that anchoring against historical evidence when extrapolating sample sizes should be adopted when such data are available. The methods are illustrated on binary and survival endpoints. 10.1002/sim.10121 http://creativecommons.org/licenses/by/4.0/ |
| title | Sample size determination for prediction models via learning‐type curves |
| topic | Statistics in Medicine |
| url | https://onlinelibrary.wiley.com/doi/10.1002/sim.10121 |