hwk3-1 - Subject-specific observed profiles of tpfbmd vs...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
tpfbmd 0.44 0.46 0.48 0.50 0.52 0.54 0.56 0.58 0.60 0.62 0.64 0.66 0.68 0.70 0.72 0.74 0.76 0.78 0.80 0.82 0.84 0.86 0.88 0.90 months 0 1 02 03 Subject-specific observed profiles of tpfbmd vs months participation in stud 0 y Blue profiles are controls, red are gymnasts tpfbmd 0.44 0.46 0.48 0.50 0.52 0.54 0.56 0.58 0.60 0.62 0.64 0.66 0.68 0.70 0.72 0.74 0.76 0.78 0.80 0.82 0.84 0.86 0.88 0.90 age 3456789 1 0 Subject-specific observed profiles of tpfbmd vs age 1 1
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Sample size in each group 1 The FREQ Procedure Cumulative Cumulative group Frequency Percent Frequency Percent ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ CON 78 54.55 78 54.55 GYM 65 45.45 143 100.00
Background image of page 2
M1: Strategy 4 analysis - RM-ANCOVA treating baseline as a covariate 2 Gives 4 df gymnast*time test Unspecified residual var-cov structure The Mixed Procedure Class Level Information Class Levels Values group 2 CON GYM months 4 6 12 18 24 id 143 1003 1004 1005 1006 1007 1008 1009 1011 1013 1017 1018 1019 1020 1021 1023 1024 1025 1028 1029 1030 1031 1034 1035 1036 1037 1039 1040 1041 1042 1043 1047 1048 1049 1051 1053 1055 1057 1058 1059 1061 1062 1063 1064 1065 1069 1072 1073 1074 1075 1076 1077 1078 1082 1084 1089 1090 1091 1094 1095 1096 1097 1098 1100 1101 1102 1103 1104 1105 1106 1107 1110 1111 1112 1113 1115 1117 1118 1119 1120 1122 1126 1127 1128 1129 1134 1135 1138 1139 1140 1141 1142 1143 1145 1146 1147 1148 1150 1152 1153 1155 1156 1157 1158 1159 1161 1162 1164 1165 1166 1168 1169 1170 1172 1173 1176 1178 1179 1180 1181 1182 1183 1185 1186 1187 1188 1189 1191 1192 1193 1194 1195 1196 1197 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 grptime 8 CON12 CON18 CON24 CON6 GYM12 GYM18 GYM24 GYM6 Convergence criteria met. Estimated R Correlation Matrix for id 1003 Row Col1 Col2 Col3 Col4 1 1.0000 0.6201 0.5803 0.5000 2 0.6201 1.0000 0.6523 0.6683 3 0.5803 0.6523 1.0000 0.7811 4 0.5000 0.6683 0.7811 1.0000
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
M1: Strategy 4 analysis - RM-ANCOVA treating baseline as a covariate 3 Gives 4 df gymnast*time test Unspecified residual var-cov structure The Mixed Procedure Covariance Parameter Estimates Standard Z Cov Parm Subject Estimate Error Value Pr Z UN(1,1) id 0.000461 0.000055 8.37 <.0001 UN(2,1) id 0.000311 0.000050 6.24 <.0001 UN(2,2) id 0.000545 0.000065 8.37 <.0001 UN(3,1) id 0.000347 0.000059 5.92 <.0001 UN(3,2) id 0.000424 0.000066 6.46 <.0001 UN(3,3) id 0.000776 0.000093 8.35 <.0001 UN(4,1) id 0.000366 0.000069 5.29 <.0001 UN(4,2) id 0.000532 0.000081 6.57 <.0001 UN(4,3) id 0.000742 0.000102 7.28 <.0001 UN(4,4) id 0.001163 0.000139 8.37 <.0001 Fit Statistics -2 Res Log Likelihood -2749.8 AIC (smaller is better) -2729.8 AICC (smaller is better) -2729.4 BIC (smaller is better) -2700.2 Type 1 Tests of Fixed Effects Num Den Effect DF DF F Value Pr > F tpfbmd0 1 140 1601.03 <.0001 months 3 137 125.83 <.0001 tpfbmd0*months 3 138 0.11 0.9550 grptime 4 137 0.25 0.9102 Type 3 Tests of Fixed Effects Num Den Effect DF DF F Value Pr > F tpfbmd0 1 140 1551.76 <.0001 months 0 . . . tpfbmd0*months 3 138 0.11 0.9518 grptime 4 137 0.25 0.9102
Background image of page 4
AIC values for variants of model M1 with a variety of var-cov structures 4 Obs Descr Value 1 AIC (smaller is better) -2729.8 2 AIC (smaller is better) -2723.7 3 AIC (smaller is better) -2692.4 4 AIC (smaller is better) -2677.3 5 AIC (smaller is better) -2700.9 6 AIC (smaller is better) -2689.9 7 AIC (smaller is better) -2683.9 8 AIC (smaller is better) -2689.9 9 AIC (smaller is better) -2656.1 10 AIC (smaller is better) -2706.2 11 AIC (smaller is better) -2658.1
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
AIC values for more variants of model M1 with a variety of var-cov structures 5 These involve different parameter values for gymnasts and controls Obs Descr Value 1 AIC (smaller is better) -2721.2 2 AIC (smaller is better) -2716.1 3 AIC (smaller is better) -2690.0 4 AIC (smaller is better) -2676.0 5 AIC (smaller is better) -2695.0 6 AIC (smaller is better) -2686.6 7 AIC (smaller is better) -2680.5 8 AIC (smaller is better) -2686.6 9 AIC (smaller is better) -2656.1 10 AIC (smaller is better) -2699.1 11 AIC (smaller is better) -2655.9
Background image of page 6
M2: Strategy 4 analysis - RM-ANCOVA treating baseline as a covariate 6 Same as M1, but drop non-signif tpfbmd0*months interaction
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 8
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 31

hwk3-1 - Subject-specific observed profiles of tpfbmd vs...

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online