TABLE 5 Percentage of Rejected Under Negatively Correlated Individual Effects and Different Levels of Individual Effects and Measurement Errors,,,,,,,,,,,
,,sP,,0,,,0.3µP,,,0,
,,sI,,0,,,0,,,0.3µI,
error ,s?,,H,Hf,Wn,H,Hf,Wn,H,Hf,Wn
C ,0, ,4.7,4.8,5.4,4.8,5,5,5,4.8,5.6
C ,0.2, ,6,6,7.2,5.8,5.2,7.2,5.8,5.4,7.8
C ,0.4, ,37,30.7,40,36.4,31.6,39.7,36.4,31.8,39.6
C ,0.6, ,96,90.8,95,95.6,91.6,94.7,96,91,94.6
B ,0, ,5,5.2,5.2,5.4,5.5,6.6,5.4,5.2,5.6
B ,0.2, ,7,6.8,8.6,8.4,7.4,10.4,6.2,6,8
B ,0.4, ,37.8,31.8,39.8,42.2,35.6,43.8,38.4,32.4,40.6
B ,0.6, ,95.7,90.3,94.4,96.8,92.8,96.2,96,92.2,95
Key: C = classical measurement errors; B = Berkson measurement errors.,,,,,,,,,,,
Note: The null hypothesis was rejected for large and small values of the test statistic at the 5% level (two-tailed test).,,,,,,,,,,,