The mysterious ρ -between or within studies [General Statistics]
Dear Nasty!
» So, for ρ=0.8 I get now
»
» 1e+06 simulations. Time consumed (secs)
» user system elapsed
» 979.75 95.18 1115.72
» Intersection null P(Type I Error) theta0 #1 theta0 #2
» 1 H_A01 n H_Ca 0.048904 0.8000000 0.953047
» 2 H_A02 n H_Ca 0.049193 1.2500000 1.056122
» 3 H_Aa n H_C01 0.049620 0.9709781 0.900000
» 4 H_Aa n H_C02 0.050024 1.0000000 1.111111
» ...
» and inflation at 0.050024
Don't forget, that type1error.2TOST() is based on simulations now.
That's mainly because our (authors team of PowerTOST) brain is to small to derive an analytical solution. And simulations have an error. F.i. in setting the seed for the random number generator.
In that light 0.050024 is of course not an alpha-inflation. As we know from theory, the IUT principle, the combination of two TOST via and assures alpha<=0.05. Thus interpret the result as 0.05. Full stop
» As I understand estimated from several datasets ρ is "between studies" ρ while in Power2TOST we need to use "within studies ρ".
I myself don't know what ρ or what value of ρ should be used. As I wrote above: It's under discussion now.
Thus if you or somebody else have an opinion or arguments: Give it to me.
» So, for ρ=0.8 I get now
»
> type1error.2TOST(CV=c(0.3, 0.25), n=122, theta1=c(0.8, 0.9),
» + theta2=c(1.25, 1/0.9), rho=0.8, details=TRUE)» 1e+06 simulations. Time consumed (secs)
» user system elapsed
» 979.75 95.18 1115.72
» Intersection null P(Type I Error) theta0 #1 theta0 #2
» 1 H_A01 n H_Ca 0.048904 0.8000000 0.953047
» 2 H_A02 n H_Ca 0.049193 1.2500000 1.056122
» 3 H_Aa n H_C01 0.049620 0.9709781 0.900000
» 4 H_Aa n H_C02 0.050024 1.0000000 1.111111
» ...
» and inflation at 0.050024
Don't forget, that type1error.2TOST() is based on simulations now.
That's mainly because our (authors team of PowerTOST) brain is to small to derive an analytical solution. And simulations have an error. F.i. in setting the seed for the random number generator.
type1error.2TOST(CV=c(0.3, 0.25), n=122, theta1=c(0.8, 0.9), theta2=c(1.25, 1/0.9), rho=0.8, setseed=F)
[K1] 0.049977
In that light 0.050024 is of course not an alpha-inflation. As we know from theory, the IUT principle, the combination of two TOST via and assures alpha<=0.05. Thus interpret the result as 0.05. Full stop
» As I understand estimated from several datasets ρ is "between studies" ρ while in Power2TOST we need to use "within studies ρ".
I myself don't know what ρ or what value of ρ should be used. As I wrote above: It's under discussion now.
Thus if you or somebody else have an opinion or arguments: Give it to me.
—
Regards,
Detlew
Regards,
Detlew
Complete thread:
- Power and type I adjustment for 222BE when several PK prams considered - Louis52, 2018-02-07 21:04 [General Statistics]
- Power and type I adjustment for 222BE when several PK prams considered - ElMaestro, 2018-02-07 22:37
- Power of two combined TOST - d_labes, 2018-02-08 14:12
- The mysterious ρ - Helmut, 2018-02-08 21:40
- ρ ρ ρ your boat - ElMaestro, 2018-02-09 09:58
- ρ ρ ρ your boat - Helmut, 2018-02-09 19:20
- ...and a bonus question - ElMaestro, 2018-02-09 11:32
- ...and a bonus question - Helmut, 2018-02-09 19:43
- If u see a BEar don't forget to R oaρ - Astea, 2018-02-09 20:47
- package install failed - d_labes, 2018-02-11 13:23
- If u see a BEar don't forget to R oaρ - Astea, 2018-02-09 20:47
- ...and a bonus question - Helmut, 2018-02-09 19:43
- The mysterious ρ -between or within studies - d_labes, 2018-02-11 15:57
- The mysterious ρ -between or within studies - Helmut, 2018-02-11 17:31
- The mysterious ρ -between or within studies - d_labes, 2018-02-11 19:21
- The mysterious ρ within studies - Helmut, 2018-02-11 22:44
- The mysterious ρ within studies - nobody, 2018-02-12 09:08
- The mysterious ρ within studies - Helmut, 2018-02-11 22:44
- The mysterious ρ -between or within studies - d_labes, 2018-02-11 19:21
- The mysterious ρ -between or within studies - Astea, 2018-02-11 17:34
- The mysterious ρ -between or within studies - d_labes, 2018-02-11 19:36
- The mysterious ρ within studies - Helmut, 2018-02-11 22:48
- The mysterious ρ -between or within studies - Helmut, 2018-02-11 17:31
- ρ ρ ρ your boat - ElMaestro, 2018-02-09 09:58
- The mysterious ρ - Helmut, 2018-02-08 21:40
- Power and type I adjustment for 222BE when several PK prams considered - ElMaestro, 2018-02-11 18:43
- Power and type I adjustment for 222BE when several PK prams considered - nobody, 2018-02-12 09:10
- reminds me of something - ElMaestro, 2018-02-12 09:18
- reminds me of something - nobody, 2018-02-12 09:38
- Hammer to fall - Astea, 2018-02-12 18:37
- Put the sickle to the corn - Helmut, 2018-02-13 01:37
- Put the sickle to the corn - Astea, 2018-02-13 08:27
- Put the sickle to the corn - Helmut, 2018-02-13 12:45
- Put the sickle to the corn - Relaxation, 2018-02-13 10:37
- multiple units, lag-times - Helmut, 2018-02-13 13:31
- Put the sickle to the corn - Astea, 2018-02-13 08:27
- Put the sickle to the corn - Helmut, 2018-02-13 01:37
- Hammer to fall - Astea, 2018-02-12 18:37
- reminds me of something - nobody, 2018-02-12 09:38
- reminds me of something - ElMaestro, 2018-02-12 09:18
- Power and type I adjustment for 222BE when several PK prams considered - nobody, 2018-02-12 09:10