Power comparison with dependent data

We set correlation equal to c(1,0.3,0) in power calculation.

  1. reg
##       0.1    0.2   0.3   0.4    0.5
## 5  0.3600 0.6600 0.700 0.840 0.8600
## 10 0.2600 0.4700 0.470 0.600 0.6300
## 15 0.2533 0.3533 0.440 0.400 0.4333
## 20 0.2450 0.4000 0.430 0.440 0.3500
## 25 0.2480 0.3320 0.336 0.404 0.3600
  1. smo
##     0.1   0.2   0.3    0.4    0.5
## 5  0.26 0.500 0.860 0.9400 0.9800
## 10 0.27 0.450 0.520 0.7400 0.8400
## 15 0.22 0.400 0.440 0.5467 0.5333
## 20 0.20 0.245 0.390 0.3750 0.4450
## 25 0.16 0.308 0.336 0.3800 0.3880
  1. difference
##         0.1      0.2   0.3     0.4   0.5
## 5  -0.10000 -0.16000  0.16  0.1000 0.120
## 10  0.01000 -0.02000  0.05  0.1400 0.210
## 15 -0.03333  0.04667  0.00  0.1467 0.100
## 20 -0.04500 -0.15500 -0.04 -0.0650 0.095
## 25 -0.08800 -0.02400  0.00 -0.0240 0.028
  1. reg
##       0.1    0.2    0.3    0.4    0.5
## 5  0.3640 0.6600 0.8080 0.8920 0.8680
## 10 0.3080 0.5240 0.5800 0.5900 0.6100
## 15 0.2960 0.4080 0.4627 0.4653 0.4640
## 20 0.2620 0.3610 0.3680 0.3980 0.4020
## 25 0.2672 0.2896 0.3552 0.3320 0.3696
  1. smo
##       0.1    0.2    0.3    0.4    0.5
## 5  0.2560 0.4480 0.6120 0.7760 0.9000
## 10 0.1820 0.2980 0.3340 0.3940 0.3880
## 15 0.1467 0.1987 0.2093 0.1853 0.1760
## 20 0.1190 0.1300 0.1130 0.1370 0.1310
## 25 0.1008 0.0920 0.0976 0.0840 0.0976
  1. difference
##        0.1     0.2     0.3    0.4    0.5
## 5  -0.1080 -0.2120 -0.1960 -0.116  0.032
## 10 -0.1260 -0.2260 -0.2460 -0.196 -0.222
## 15 -0.1493 -0.2093 -0.2533 -0.280 -0.288
## 20 -0.1430 -0.2310 -0.2550 -0.261 -0.271
## 25 -0.1664 -0.1976 -0.2576 -0.248 -0.272

The differences of power between two methods are similar to previous calculation result. The power decreases if the data not independent.