## Exact Power [R for BE/BA]

Dear all,

I made an investment and was able to compare results from

nQuery requires sqrt(MSE) rather than CV as input = sqrt(ln(CV²+1)). The maximum precision is 6 decimal places. Therefore 0.1980422 rather than the value in full precision (0.1980422004353650284627675024887) like in

Values agree nicely with the exception of a small discrepancy at GMR=1 (83.31980%

If I recall it correctly Dieter Hauschke wrote something about numerical problems in power estimation at T/R=1 in one of his papers.

I love validating software.

I made an investment and was able to compare results from

**R**with the current version 7 of nQuery Advisor:*```
+======+=====+===========+========+==========+
```

| GMR | n | R-code | nQuery Advisor 7 |

+------+-----+-----------+--------+----------+

| 0.85 | 134 | 0.8017723 | 80.17 | 80.17722 |

| 0.90 | 38 | 0.8154940 | 81.54 | 81.54941 |

| 0.95 | 20 | 0.8346802 | 83.46 | 83.46802 |

| 1.00 | 16 | 0.8331980 | 83.32 | 83.32001 |

| 1.05 | 18 | 0.8001854 | 80.01 | 80.01854 |

| 1.10 | 32 | 0.8100682 | 81.00 | 81.00682 |

| 1.15 | 72 | 0.8042181 | 80.42 | 80.42180 |

| 1.20 | 294 | 0.8019247 | 80.19 | 80.19246 |

+======+=====+===========+========+==========+

nQuery requires sqrt(MSE) rather than CV as input = sqrt(ln(CV²+1)). The maximum precision is 6 decimal places. Therefore 0.1980422 rather than the value in full precision (0.1980422004353650284627675024887) like in

**R**is used. But you see this value only if clicking the respective cell; both displayed and printed is 0.198042. nQuery gives power in percent with a maximum number of two decimal digits. That's a pity and values in the first column were the best one could get (display/print). If clicking in each cell power is given in percent to 5 decimal digits (7 significant digits); second column. It's clear that results are not rounded, but truncated. This looked weird first, but at least for power it makes sense.Values agree nicely with the exception of a small discrepancy at GMR=1 (83.31980%

*vs.*83.32001%).If I recall it correctly Dieter Hauschke wrote something about numerical problems in power estimation at T/R=1 in one of his papers.

I love validating software.

**Janet D Elashoff**

*nQuery Advisor Version 7.0*

Statistical Solutions, Cork, Ireland (2007)

—

Cheers,

Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. ☼

Science Quotes

Cheers,

Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. ☼

Science Quotes

### Complete thread:

- R-Code for Power in 2×2 Cross-over Helmut 2006-12-28 21:33
- Power in Diletti's sample size table Helmut 2007-01-01 23:59
- Approximate Power Helmut 2007-01-06 15:17
- Power Calculation for replicate crossover design mathews 2008-04-03 13:19
- Power Calculation for replicate crossover design Helmut 2008-04-03 13:49
- Power Calculation for replicate crossover design mathews 2008-04-04 13:21

- Power Calculation for replicate crossover design Helmut 2008-04-03 13:49
- Exact Power Helmut 2009-09-25 19:20
- Exact Power ElMaestro 2009-09-27 23:27
- Software validation Helmut 2009-09-28 15:21
- All the important answers are here! ElMaestro 2009-09-28 19:12
- All the important answers are here! Helmut 2009-09-28 19:52
- All the important answers are here! ElMaestro 2009-09-28 20:18
- All the important answers are here! yjlee168 2009-09-28 21:30
- All the important answers are here! Helmut 2009-09-28 23:18
- 42! SASophylistic power oracle d_labes 2009-09-29 13:37
- Al Gore Rhythms Helmut 2009-09-29 14:52
- Al Gore Rhythms ElMaestro 2009-09-29 18:14
- Al Gore Rhythms yjlee168 2009-09-29 23:55

- Al Gore Rhythms Helmut 2009-09-29 14:52

- 42! SASophylistic power oracle d_labes 2009-09-29 13:37
- All the important answers are here! yjlee168 2009-09-29 23:49

- All the important answers are here! Helmut 2009-09-28 23:18

- All the important answers are here! Helmut 2009-09-28 19:52

- All the important answers are here! ElMaestro 2009-09-28 19:12

- Software validation Helmut 2009-09-28 15:21

- Exact Power ElMaestro 2009-09-27 23:27

- Power Calculation for replicate crossover design mathews 2008-04-03 13:19

- Approximate Power Helmut 2007-01-06 15:17
- R-Code for Power in 2×2 Cross-over KDA 2019-01-23 04:24
- R-Code for Power in 2×2 Cross-over Ohlbe 2019-01-27 18:24

- Power in Diletti's sample size table Helmut 2007-01-01 23:59