### Abstract

In designing scientific experiments, power analysis is too often given a superficial treatment - choice of sample size is often made based on idealized distributions and simplistic tests that do not reflect the real-world constraints under which the actual data will be collected. We have developed a general Monte Carlo framework for two-group comparisons which samples points from a two-dimensional parameter space and at each point generates simulated datasets which are compared to simulated datasets for a 'control group' at a fixed point in the parameter space. Rather than uniformly sampling this parameter space, our algorithm rapidly converges on a contour corresponding to the smallest detectable difference for the sample size of interest. We apply this framework, implemented as an R library called PowerTrip, to directly comparing the performance and sensitivity to sample size of the Gompertz survival model to several other commonly used survival models. We find that the Gompertz mortality model performs approximately as well as the Weibull and the Cox models throughout most of the parameter space, but outperforms the competing models in cases where initial mortality rate (IMR) and rate of acceleration (RoA) change in opposite directions.

Original language | English (US) |
---|---|

Title of host publication | 2017 IEEE Symposium on Computers and Communications, ISCC 2017 |

Publisher | Institute of Electrical and Electronics Engineers Inc. |

Pages | 290-295 |

Number of pages | 6 |

ISBN (Electronic) | 9781538616291 |

DOIs | |

State | Published - Sep 1 2017 |

Event | 2017 IEEE Symposium on Computers and Communications, ISCC 2017 - Heraklion, Greece Duration: Jul 3 2017 → Jul 7 2017 |

### Other

Other | 2017 IEEE Symposium on Computers and Communications, ISCC 2017 |
---|---|

Country | Greece |

City | Heraklion |

Period | 7/3/17 → 7/7/17 |

### Fingerprint

### Keywords

- Maximum likelihood estimation
- Monte Carlo methods
- Statistics

### ASJC Scopus subject areas

- Software
- Signal Processing
- Mathematics(all)
- Computer Science Applications
- Computer Networks and Communications

### Cite this

*2017 IEEE Symposium on Computers and Communications, ISCC 2017*(pp. 290-295). [8024544] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISCC.2017.8024544

**Biologically relevant simulations for validating risk models under small-sample conditions.** / Bokov, Alex F.; Manuel, Laura S.; Tirado-Ramos, Alfredo; Gelfond, Jonathan A; Pletcher, Scott D.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*2017 IEEE Symposium on Computers and Communications, ISCC 2017.*, 8024544, Institute of Electrical and Electronics Engineers Inc., pp. 290-295, 2017 IEEE Symposium on Computers and Communications, ISCC 2017, Heraklion, Greece, 7/3/17. https://doi.org/10.1109/ISCC.2017.8024544

}

TY - GEN

T1 - Biologically relevant simulations for validating risk models under small-sample conditions

AU - Bokov, Alex F.

AU - Manuel, Laura S.

AU - Tirado-Ramos, Alfredo

AU - Gelfond, Jonathan A

AU - Pletcher, Scott D.

PY - 2017/9/1

Y1 - 2017/9/1

N2 - In designing scientific experiments, power analysis is too often given a superficial treatment - choice of sample size is often made based on idealized distributions and simplistic tests that do not reflect the real-world constraints under which the actual data will be collected. We have developed a general Monte Carlo framework for two-group comparisons which samples points from a two-dimensional parameter space and at each point generates simulated datasets which are compared to simulated datasets for a 'control group' at a fixed point in the parameter space. Rather than uniformly sampling this parameter space, our algorithm rapidly converges on a contour corresponding to the smallest detectable difference for the sample size of interest. We apply this framework, implemented as an R library called PowerTrip, to directly comparing the performance and sensitivity to sample size of the Gompertz survival model to several other commonly used survival models. We find that the Gompertz mortality model performs approximately as well as the Weibull and the Cox models throughout most of the parameter space, but outperforms the competing models in cases where initial mortality rate (IMR) and rate of acceleration (RoA) change in opposite directions.

AB - In designing scientific experiments, power analysis is too often given a superficial treatment - choice of sample size is often made based on idealized distributions and simplistic tests that do not reflect the real-world constraints under which the actual data will be collected. We have developed a general Monte Carlo framework for two-group comparisons which samples points from a two-dimensional parameter space and at each point generates simulated datasets which are compared to simulated datasets for a 'control group' at a fixed point in the parameter space. Rather than uniformly sampling this parameter space, our algorithm rapidly converges on a contour corresponding to the smallest detectable difference for the sample size of interest. We apply this framework, implemented as an R library called PowerTrip, to directly comparing the performance and sensitivity to sample size of the Gompertz survival model to several other commonly used survival models. We find that the Gompertz mortality model performs approximately as well as the Weibull and the Cox models throughout most of the parameter space, but outperforms the competing models in cases where initial mortality rate (IMR) and rate of acceleration (RoA) change in opposite directions.

KW - Maximum likelihood estimation

KW - Monte Carlo methods

KW - Statistics

UR - http://www.scopus.com/inward/record.url?scp=85030531050&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85030531050&partnerID=8YFLogxK

U2 - 10.1109/ISCC.2017.8024544

DO - 10.1109/ISCC.2017.8024544

M3 - Conference contribution

AN - SCOPUS:85030531050

SP - 290

EP - 295

BT - 2017 IEEE Symposium on Computers and Communications, ISCC 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -