tailieunhanh - A Methodology for the Health Sciences - part 7

Các giới hạn này rộng lớn hơn nhiều so với những người dựa trên sự so sánh mỗi t-Thống kê. Điều này là do mục đích duy nhất là thay thế của t29, 0,975 = 2,05 5F5, 29, = 3,57. Do đó, chiều rộng khoảng tin cậy được tăng lên bởi một yếu tố của 1,74 hoặc 74% =. | SIMULTANEOUS CONFIDENCE INTERVALS AND TESTS FOR LINEAR MODELS 527 or bi sbị i 1 2 . 5 For the regression coefficient of DPMB the interval is resulting in 95 confidence limits of . Computing these values the confidence intervals are as follows Variable Limits Variable Limits Lower Upper Lower Upper DPMB Time Trauma Lymph A Lymph B These limits are much wider than those based on a per comparison t-statistic. This is due solely to the replacement of t29 by ự5F5 29 . Hence the confidence interval width is increased by a factor of or 74 . Example . In a one-way anova situation using the notation of Section if we wish simultaneous confidence intervals for all I means then d I m and the standard error of the estimate of U is Thus the confidence intervals are of the form - 7T------- MSe Y 7 IF .- 1-. i 1 . . . I Suppose that we want simultaneous 99 confidence intervals for the morphine binding data of Problem . The confidence interval for the chronic group is 1 4 422 y F4 24 18 or The four simultaneous 99 confidence intervals are Limits Limits Group Lower Upper Group Lower Upper 4 Chronic n Dialysis 19 Acute 74 Anephric 528 MULTIPLE COMPARISONS As all four intervals overlap we cannot conclude immediately from this approach that the means differ at the level . To compare two means we can also consider confidence intervals for I ni .As the Scheffe method allows us to look at all linear combinations we may also consider the confidence interval for I i i. The formula for the simultaneous confidence intervals is Yi. Yi-. JIF . 1 L MSA x x V ni nv i i 1 . I i i In this case the confidence intervals are Limits Limits Contrast Lower Upper Contrast Lower Upper 1 2 U 3 3 F-1 13 u 4 1 4 3 3 4 As the interval