{"id":444,"date":"2019-05-05T21:15:40","date_gmt":"2019-05-05T12:15:40","guid":{"rendered":"https:\/\/best-biostatistics.com\/toukei-er\/entry\/bland-altman-analysis-in-r\/"},"modified":"2024-10-10T23:12:14","modified_gmt":"2024-10-10T14:12:14","slug":"bland-altman-analysis-in-r","status":"publish","type":"post","link":"https:\/\/best-biostatistics.com\/toukei-er\/entry\/bland-altman-analysis-in-r\/","title":{"rendered":"R \u3067\u30d6\u30e9\u30f3\u30c9 \u30a2\u30eb\u30c8\u30de\u30f3 \u5206\u6790\u3092\u884c\u3046\u65b9\u6cd5"},"content":{"rendered":"\n

\u30d6\u30e9\u30f3\u30c9 \u30a2\u30eb\u30c8\u30de\u30f3 \u5206\u6790\u306f\u3001\u4e8c\u3064\u306e\u6e2c\u5b9a\u7cfb\u306e\u7d50\u679c\u304c\u4e00\u81f4\u3057\u3066\u3044\u308b\u304b\u3069\u3046\u304b\u3092\u78ba\u8a8d\u3059\u308b\u65b9\u6cd5\u3002<\/p>\n\n\n\n

\u30d6\u30e9\u30f3\u30c9 \u30a2\u30eb\u30c8\u30de\u30f3 \u30d7\u30ed\u30c3\u30c8\u306b\u3001\u56de\u5e30\u76f4\u7dda\u3092\u5408\u308f\u305b\u308b\u3068\u4e0d\u4e00\u81f4\u306b\u50be\u5411\u304c\u306a\u3044\u304b\u3069\u3046\u304b\u78ba\u8a8d\u3067\u304d\u308b\u3002<\/p>\n\n\n\n\n\n\n\n

\u30d6\u30e9\u30f3\u30c9\u30a2\u30eb\u30c8\u30de\u30f3\u5206\u6790\u306e\u6e96\u5099<\/h2>\n\n\n\n

\u30d6\u30e9\u30f3\u30c9\u30a2\u30eb\u30c8\u30de\u30f3\u30d7\u30ed\u30c3\u30c8\u306f\u30012\u3064\u306e\u6e2c\u5b9a\u5024\u306e\u5e73\u5747\u5024\u3092X\u8ef8\u306b\u3001\u5dee\u3092Y\u8ef8\u306b\u3057\u3066\u30d7\u30ed\u30c3\u30c8\u3059\u308b\u3002<\/p>\n\n\n\n

blandr \u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u4f7f\u3046\u3068\u7c21\u5358\u306b\u63cf\u3051\u308b\u3002<\/p>\n\n\n\n

\u307e\u305a blandr \u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b\u3002<\/p>\n\n\n\n

install.packages<\/span>(<\/span>\"blandr\"<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u304c\u7d42\u4e86\u3057\u305f\u3089\u547c\u3073\u51fa\u3059\u3002<\/p>\n\n\n\n
\u3053\u308c\u3067\u6e96\u5099\u5b8c\u4e86\u3002<\/p>\n\n\n\n
library<\/span>(<\/span>blandr)<\/span>\n<\/code><\/pre>\n\n\n\n
Bland and Altman 1986 \u8ad6\u6587<\/a>\u304b\u3089PEFR (Peak Expiratory Flow Rate)\u306e\u30c7\u30fc\u30bf\u3092\u4f7f\u3063\u3066\u30b0\u30e9\u30d5\u3092\u63cf\u304f\u3002<\/p>\n\n\n\n
bland.altman.PEFR.1986\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u306e\u4e00\u5217\u76ee\uff08Wright Meter \u306e\u6e2c\u5b9a\u4e00\u56de\u76ee\uff09\u3068\u4e09\u5217\u76ee\uff08Mini Wright Meter \u306e\u6e2c\u5b9a\u4e00\u56de\u76ee\uff09\u3092\u4f7f\u3046\u3002<\/p>\n\n\n\n
dat0 <-<\/span> bland.altman.PEFR.1986\ndat <-<\/span> blandr.data.preparation<\/span>(<\/span>dat0[,<\/span>1<\/span>],<\/span>dat0[,<\/span>3<\/span>],<\/span>sig.level=<\/span>0.95<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
dat\u306e\u5185\u5bb9\u3092summary(dat)\u3067\u898b\u3066\u307f\u308b\u3002<\/p>\n\n\n\n
\u4e8c\u3064\u306e\u6e2c\u5b9a\u65b9\u6cd5\u306fmethod1\u3068method2\u3068\u3044\u3046\u540d\u524d\u306b\u306a\u308b\u3002<\/p>\n\n\n\n
><\/span> summary<\/span>(<\/span>dat)<\/span>\nmethod1         method2\nMin.   :<\/span>178.0<\/span>   Min.   :<\/span>259.0<\/span>\n1st Qu.:<\/span>417.0<\/span>   1st Qu.:<\/span>380.0<\/span>\nMedian :<\/span>434.0<\/span>   Median :<\/span>445.0<\/span>\nMean   :<\/span>450.4<\/span>   Mean   :<\/span>452.5<\/span>\n3rd Qu.:<\/span>494.0<\/span>   3rd Qu.:<\/span>512.0<\/span>\nMax.   :<\/span>656.0<\/span>   Max.   :<\/span>658.0<\/span>\n<\/code><\/pre>\n\n\n\n
\u30d6\u30e9\u30f3\u30c9\u30a2\u30eb\u30c8\u30de\u30f3\u30d7\u30ed\u30c3\u30c8\u306e\u66f8\u304d\u65b9<\/h2>\n\n\n\n
\u30d6\u30e9\u30f3\u30c9\u30a2\u30eb\u30c8\u30de\u30f3\u30d7\u30ed\u30c3\u30c8\u306f\u4ee5\u4e0b\u306e\u3088\u3046\u306b\u8a18\u8ff0\u3059\u308b\u3068\u66f8\u3051\u308b\u3002<\/p>\n\n\n\n
with<\/span>(<\/span>dat,<\/span> blandr.draw<\/span>(<\/span>method1,<\/span> method2,<\/span>\nciShading=<\/span>TRUE<\/span>,<\/span>\nplotTitle=<\/span>\"Difference against mean for PEFR data\"<\/span>))<\/span>\n<\/code><\/pre>\n\n\n\n
\u30ad\u30e2\u306f blandr.draw(method1, method2)<\/p>\n\n\n\n
ciShading=TRUE\u3067Bias\u306e\u7bc4\u56f2\uff08\u9752\uff09\u3068Lower limit of agreement\u306e\u7bc4\u56f2\uff08\u8d64\uff09\u3001Upper limit of agreement\u306e\u7bc4\u56f2\uff08\u7dd1\uff09\u304c\u5857\u3089\u308c\u308b\u3002<\/p>\n\n\n\n
Limit of agreement\u3088\u308a\u3082\u96e2\u308c\u305f\u70b9\u304c\u306a\u3044\u304b\u3069\u3046\u304b\u3001\u96e2\u308c\u305f\u70b9\u304c\u591a\u3044\u304b\u3069\u3046\u304b\u3067\u3001\u4e00\u81f4\u3057\u3066\u3044\u308b\u5177\u5408\u3092\u5224\u65ad\u3059\u308b\u3002<\/p>\n\n\n\n
\u305d\u306e\u969b\u3001Limit of agreement\u306b\u3082\u5e45\u304c\u3042\u308b\u3053\u3068\u3082\u8003\u616e\u306b\u5165\u308c\u308b\u3002<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
X \u8ef8 Y \u8ef8\u3092\u5909\u66f4\u3059\u308b\u306b\u306f\u4ee5\u4e0b\u306e\u3088\u3046\u306b\u3059\u308b<\/p>\n\n\n\n
\u307e\u305a\u3001\u57fa\u672c\u7684\u306a\u30d7\u30ed\u30c3\u30c8\u3092\u30aa\u30d6\u30b8\u30a7\u30af\u30c8\u306b\u3059\u308b<\/p>\n\n\n\n
\u305d\u3053\u306b X \u8ef8 Y \u8ef8\u306e\u9650\u754c\u5024\u3092\uff0b\u3067\u3064\u306a\u304e\u5408\u308f\u305b\u3066\u6307\u5b9a\u3057\u3001\u63cf\u753b\u3059\u308b\uff08coord_cartesian \u3092\u4f7f\u7528\u3059\u308b\uff09<\/p>\n\n\n\n
\u30d6\u30e9\u30f3\u30c9\u30a2\u30eb\u30c8\u30de\u30f3\u30d7\u30ed\u30c3\u30c8\u81ea\u4f53\u306f\u3001ggplot2 \u3092\u7528\u3044\u3066\u3044\u308b\u306e\u3067\u3001\u305d\u306e\u8a18\u6cd5\u306b\u306e\u3063\u3068\u308b\u5fc5\u8981\u304c\u3042\u308b<\/p>\n\n\n\n
library<\/span>(<\/span>ggplot2)<\/span>\n# \u57fa\u672c\u7684\u306a\u30d7\u30ed\u30c3\u30c8\u3092\u30aa\u30d6\u30b8\u30a7\u30af\u30c8\u306b\u3059\u308b<\/span>\nbland.altman.plot <-<\/span> with<\/span>(<\/span>dat,<\/span> blandr.draw<\/span>(<\/span>method1,<\/span> method2,<\/span>\nciShading=<\/span>TRUE<\/span>,<\/span>\nplotTitle=<\/span>'Difference against meanfor PEFR data'<\/span>))<\/span>\n# X \u8ef8 Y \u8ef8\u306e\u9650\u754c\u5024\u3092\uff0b\u3067\u6307\u5b9a\u3057\u3066\u63cf\u753b\u3059\u308b<\/span>\nbland.altman.plot +<\/span> coord_cartesian<\/span>(<\/span>xlim=<\/span>c<\/span>(<\/span>0<\/span>,<\/span>800<\/span>),<\/span> ylim=<\/span>c<\/span>(<\/span>-<\/span>200<\/span>,<\/span> 200<\/span>))<\/span>\n<\/code><\/pre>\n\n\n\n
\u305d\u3046\u3059\u308b\u3068\u3001\u4ee5\u4e0b\u306e\u3088\u3046\u306b X \u8ef8 Y \u8ef8\u306e\u7bc4\u56f2\u3092\u5909\u66f4\u3067\u304d\u308b<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
Bias \u306e\u7bc4\u56f2\u3084\u3001Limit of Agreement \u306e\u7bc4\u56f2\u3092\u5857\u308a\u3064\u3076\u3057\u3092\u53d6\u308a\u9664\u304f\u306b\u306f\u3001ciShading=FALSE \u3068\u3059\u308b\u3002<\/p>\n\n\n\n
with<\/span>(<\/span>dat,<\/span> blandr.draw<\/span>(<\/span>method1,<\/span> method2,<\/span> ciShading=<\/span>FALSE<\/span>))<\/span>\n<\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
Bias \u306e\u7bc4\u56f2\u306e\u4e0a\u9650\u30fb\u4e0b\u9650\u3001Limit of Agreement \u306e\u4e0a\u9650\u30fb\u4e0b\u9650\u306e\u7dda\u3092\u53d6\u308a\u9664\u304f\u306b\u306f\u3001ciDisplay=FALSE \u3068\u3059\u308b\u3002<\/p>\n\n\n\n
with<\/span>(<\/span>dat,<\/span> blandr.draw<\/span>(<\/span>method1,<\/span> method2,<\/span> ciShading=<\/span>FALSE<\/span>,<\/span> ciDisplay=<\/span>FALSE<\/span>))<\/span>\n<\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
\u80cc\u666f\u306e\u30b0\u30ea\u30c3\u30c9\u3092\u524a\u9664\u3059\u308b\u306b\u306f\u3001plotter=”rplot” \u3068\u3059\u308b\u3002<\/p>\n\n\n\n
with<\/span>(<\/span>dat,<\/span> blandr.draw<\/span>(<\/span>method1,<\/span> method2,<\/span> ciShading=<\/span>FALSE<\/span>,<\/span> plotter=<\/span>\"rplot\"<\/span>))<\/span>\n<\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
\uff1e\uff1e\u3082\u3046\u7d71\u8a08\u3067\u60a9\u3080\u306e\u306f\u7d42\u308f\u308a\u306b\u3057\u307e\u305b\u3093\u304b\uff1f\u00a0<\/a><\/strong><\/span><\/p>\r\n\"\"<\/a>\r\n
\u21911\u4e07\u4eba\u4ee5\u4e0a\u306e\u533b\u7642\u5f93\u4e8b\u8005\u304c\u8cfc\u8aad\u4e2d<\/span><\/strong><\/span><\/p><\/div>
\u30d6\u30e9\u30f3\u30c9\u30a2\u30eb\u30c8\u30de\u30f3\u5206\u6790\u306e\u7d50\u679c\u51fa\u529b\u65b9\u6cd5<\/h2>\n\n\n\n
blandr.display.and.draw() \u3092\u4f7f\u3046\u3068\u3001\u30b0\u30e9\u30d5\u306e\u63cf\u753b\u3068\u3068\u3082\u306b\u5206\u6790\u7d50\u679c\u3092\u8868\u793a\u3057\u3066\u304f\u308c\u308b\u3002<\/p>\n\n\n\n
\u307e\u305f\u3001blandr.output.text() \u306f\u30c6\u30ad\u30b9\u30c8\u51fa\u529b\u3060\u3051\u884c\u3046\u3002<\/p>\n\n\n\n
with<\/span>(<\/span>dat,<\/span> blandr.display.and.draw<\/span>(<\/span>method1,<\/span> method2,<\/span>\nciShading=<\/span>TRUE<\/span>,<\/span>\nplotTitle=<\/span>\"Difference against mean for PEFR data\"<\/span>))<\/span>\nwith<\/span>(<\/span>dat,<\/span> blandr.output.text<\/span>(<\/span>method1,<\/span> method2))<\/span>\n<\/code><\/pre>\n\n\n\n
Bias\u304c2\u3064\u306e\u6e2c\u5b9a\u65b9\u6cd5\u306e\u5dee\u306e\u5e73\u5747\u5024\u3002<\/p>\n\n\n\n
\u305d\u306e95%\u306e\u5206\u5e03\u7bc4\u56f2\uff08\u00b11.96SD\uff09\u304cLimits of agreement\uff08LOA\uff09\u3002<\/p>\n\n\n\n
\u4e0a\u9650\u5024\u304cUpper LOA\uff08ULoA\uff09\u3001\u4e0b\u9650\u5024\u304cLower LOA\uff08LLoA\uff09\u3002<\/p>\n\n\n\n
Bias\u3001Upper LOA\u3001Lower LOA\u305d\u308c\u305e\u308c\u306e95%\u4fe1\u983c\u533a\u9593\u304c\u8a08\u7b97\u3055\u308c\u3066\u3044\u308b\u3002<\/p>\n\n\n\n
Number of comparisons:<\/span>  17<\/span>\nMaximum value for<\/span> average measures:<\/span>  654<\/span>\nMinimum value for<\/span> average measures:<\/span>  218.5<\/span>\nMaximum value for<\/span> difference in<\/span> measures:<\/span>  73<\/span>\nMinimum value for<\/span> difference in<\/span> measures:<\/span>  -<\/span>81<\/span>\nBias:<\/span>  -<\/span>2.117647<\/span>\nStandard deviation of bias:<\/span>  38.76513<\/span>\nStandard error of bias:<\/span>  9.401925<\/span>\nStandard error for<\/span> limits of agreement:<\/span>  16.39491<\/span>\nBias:<\/span>  -<\/span>2.117647<\/span>\nBias-<\/span> upper 95<\/span>% CI:<\/span>  17.81354<\/span>\nBias-<\/span> lower 95<\/span>% CI:<\/span>  -<\/span>22.04884<\/span>\nUpper limit of agreement:<\/span>  73.86201<\/span>\nUpper LOA-<\/span> upper 95<\/span>% CI:<\/span>  108.6177<\/span>\nUpper LOA-<\/span> lower 95<\/span>% CI:<\/span>  39.10636<\/span>\nLower limit of agreement:<\/span>  -<\/span>78.0973<\/span>\nLower LOA-<\/span> upper 95<\/span>% CI:<\/span>  -<\/span>43.34165<\/span>\nLower LOA-<\/span> lower 95<\/span>% CI:<\/span>  -<\/span>112.8529<\/span>\nDerived measures:<\/span>\nMean of differences\/<\/span>means:<\/span>  -<\/span>1.158314<\/span>\nPoint estimate of bias as proportion of lowest average:<\/span>  -<\/span>0.9691749<\/span>\nPoint estimate of bias as proportion of highest average -<\/span>0.3237992<\/span>\nSpread of data between lower and upper LoAs:<\/span>  151.9593<\/span>\nBias as proportion of LoA spread:<\/span>  -<\/span>1.393562<\/span>\nBias:<\/span>\n-<\/span>2.117647  <\/span>(<\/span> -<\/span>22.04884<\/span>  to  17.81354<\/span> )<\/span>\nULoA:<\/span>\n73.86201  <\/span>(<\/span> 39.10636<\/span>  to  108.6177<\/span> )<\/span>\nLLoA:<\/span>\n-<\/span>78.0973  <\/span>(<\/span> -<\/span>112.8529<\/span>  to  -<\/span>43.34165<\/span> )<\/span>\n<\/code><\/pre>\n\n\n\n
blandr.statistics()\u3067\u5168\u7d71\u8a08\u5024\u304c\u51fa\u529b\u3055\u308c\u308b\u3002<\/p>\n\n\n\n
><\/span> (<\/span>ba.stats <-<\/span> with<\/span>(<\/span>dat,<\/span> blandr.statistics<\/span>(<\/span>method1,<\/span> method2)))<\/span>\n$<\/span>means\n[<\/span>1<\/span>]<\/span> 503.0<\/span> 412.5<\/span> 518.0<\/span> 431.0<\/span> 488.0<\/span> 578.5<\/span> 388.5<\/span> 411.0<\/span> 654.0<\/span> 439.0<\/span> 424.5<\/span> 641.0<\/span> 263.5<\/span>\n[<\/span>14<\/span>]<\/span> 477.5<\/span> 218.5<\/span> 386.5<\/span> 439.0<\/span>\n$<\/span>differences\n[<\/span>1<\/span>]<\/span> -<\/span>18<\/span> -<\/span>35<\/span>  -<\/span>4<\/span>   6<\/span> -<\/span>24<\/span> -<\/span>43<\/span>  49<\/span>  62<\/span>  -<\/span>8<\/span> -<\/span>12<\/span> -<\/span>15<\/span>  30<\/span>   7<\/span>   1<\/span> -<\/span>81<\/span>  73<\/span> -<\/span>24<\/span>\n$<\/span>method1\n[<\/span>1<\/span>]<\/span> 494<\/span> 395<\/span> 516<\/span> 434<\/span> 476<\/span> 557<\/span> 413<\/span> 442<\/span> 650<\/span> 433<\/span> 417<\/span> 656<\/span> 267<\/span> 478<\/span> 178<\/span> 423<\/span> 427<\/span>\n$<\/span>method2\n[<\/span>1<\/span>]<\/span> 512<\/span> 430<\/span> 520<\/span> 428<\/span> 500<\/span> 600<\/span> 364<\/span> 380<\/span> 658<\/span> 445<\/span> 432<\/span> 626<\/span> 260<\/span> 477<\/span> 259<\/span> 350<\/span> 451<\/span>\n$<\/span>sig.level\n[<\/span>1<\/span>]<\/span> 0.95<\/span>\n$<\/span>sig.level.convert.to.z\n[<\/span>1<\/span>]<\/span> 1.959964<\/span>\n$<\/span>bias\n[<\/span>1<\/span>]<\/span> -<\/span>2.117647<\/span>\n$<\/span>biasUpperCI\n[<\/span>1<\/span>]<\/span> 17.81354<\/span>\n$<\/span>biasLowerCI\n[<\/span>1<\/span>]<\/span> -<\/span>22.04884<\/span>\n$<\/span>biasStdDev\n[<\/span>1<\/span>]<\/span> 38.76513<\/span>\n$<\/span>biasSEM\n[<\/span>1<\/span>]<\/span> 9.401925<\/span>\n$<\/span>LOA_SEM\n[<\/span>1<\/span>]<\/span> 16.39491<\/span>\n$<\/span>upperLOA\n[<\/span>1<\/span>]<\/span> 73.86201<\/span>\n$<\/span>upperLOA_upperCI\n[<\/span>1<\/span>]<\/span> 108.6177<\/span>\n$<\/span>upperLOA_lowerCI\n[<\/span>1<\/span>]<\/span> 39.10636<\/span>\n$<\/span>lowerLOA\n[<\/span>1<\/span>]<\/span> -<\/span>78.0973<\/span>\n$<\/span>lowerLOA_upperCI\n[<\/span>1<\/span>]<\/span> -<\/span>43.34165<\/span>\n$<\/span>lowerLOA_lowerCI\n[<\/span>1<\/span>]<\/span> -<\/span>112.8529<\/span>\n$<\/span>proportion\n[<\/span>1<\/span>]<\/span>  -<\/span>3.5785288<\/span>  -<\/span>8.4848485<\/span>  -<\/span>0.7722008<\/span>   1.3921114<\/span>  -<\/span>4.9180328<\/span>  -<\/span>7.4330164<\/span>\n[<\/span>7<\/span>]<\/span>  12.6126126<\/span>  15.0851582<\/span>  -<\/span>1.2232416<\/span>  -<\/span>2.7334852<\/span>  -<\/span>3.5335689<\/span>   4.6801872<\/span>\n[<\/span>13<\/span>]<\/span>   2.6565465<\/span>   0.2094241<\/span> -<\/span>37.0709382<\/span>  18.8874515<\/span>  -<\/span>5.4669704<\/span>\n$<\/span>no.of.observations\n[<\/span>1<\/span>]<\/span> 17<\/span>\n$<\/span>regression.equation\n[<\/span>1<\/span>]<\/span> \"y(differences) = 0.029 x(means) + -15\"<\/span>\n$<\/span>regression.fixed.slope\nmeans\n0.029<\/span>\n$<\/span>regression.fixed.intercept\n(<\/span>Intercept)<\/span>\n-<\/span>15<\/span>\n<\/code><\/pre>\n\n\n\n
$regression.equation\u306f\u3001\u63a8\u5b9a\u56de\u5e30\u76f4\u7dda\u3067\u3001\u5207\u7247\uff08intercept: $regression.fixed.intercept\uff09\u3068\u50be\u304d\uff08slope: $regression.fixed.slope\uff09\u304c\u8a08\u7b97\u3055\u308c\u3066\u3044\u308b\u3002<\/p>\n\n\n\n
\u3053\u308c\u3092\u4f7f\u3063\u3066method1\u3068method2\u306e\u5e73\u5747\u3068\u5dee\u306e\u56de\u5e30\u76f4\u7dda\u3092\u5148\u307b\u3069\u306e\u30b0\u30e9\u30d5\u306b\u4e57\u305b\u308b\u3068\u4ee5\u4e0b\u306e\u3088\u3046\u306b\u306a\u308b\u3002<\/p>\n\n\n\n
abline<\/span>(<\/span>a=<\/span>ba.stats$<\/span>regression.fixed.intercept,<\/span>b=<\/span>ba.stats$<\/span>regression.fixed.slope,<\/span>lwd=<\/span>2<\/span>)<\/span>\ntext<\/span>(<\/span>400<\/span>,<\/span>20<\/span>,<\/span>ba.stats$<\/span>regression.equation)<\/span>\n<\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
\u56de\u5e30\u5206\u6790\u306e\u7d50\u679c\u304b\u3089\u3082\u308f\u304b\u308b\u901a\u308a\u3001\u50be\u304d\u306f\u9650\u308a\u306a\u304f\u30bc\u30ed\u306b\u8fd1\u304f\u3001\u5e73\u5747\u306e\u5927\u5c0f\u3067\u3001\u5dee\u306e\u5927\u5c0f\u304c\u6c7a\u307e\u308b\u3068\u3044\u3046\u3088\u3046\u306a\u50be\u5411\u306f\u307f\u3089\u308c\u306a\u3044\u3002<\/p>\n\n\n\n
\u3057\u305f\u304c\u3063\u3066\u3001\u7cfb\u7d71\u8aa4\u5dee\uff08\u4e00\u5b9a\u306e\u50be\u5411\u3092\u6301\u3063\u305f\u30ba\u30ec\uff09\u306f\u306a\u3044\u3068\u8a00\u3048\u308b\u3002<\/p>\n\n\n\n
\u56de\u5e30\u5206\u6790\u306e\u524d\u306b\u5e73\u5747mean\u3068\u5deedifference\u3092\u4f5c\u6210\u3057\u3066\u304a\u304f\u3002<\/p>\n\n\n\n
dat$<\/span>difference <-<\/span> dat$<\/span>method1 -<\/span> dat$<\/span>method2\ndat$<\/span>mean <-<\/span> (<\/span>dat$<\/span>method1+<\/span>dat$<\/span>method2)<\/span>\/<\/span>2<\/span>\n<\/code><\/pre>\n\n\n\n
mean\u306eEstimate\u306f\u3001p=0.749\u3067\u50be\u304d\u304c\u30bc\u30ed\u306e\u5e30\u7121\u4eee\u8aac\u304c\u68c4\u5374\u3067\u304d\u305a\u3001\u50be\u304d\u30bc\u30ed\u3067\u306a\u3044\u3068\u306f\u8a00\u3048\u306a\u3044\u3002<\/p>\n\n\n\n
><\/span> dat$<\/span>difference <-<\/span> dat$<\/span>method1 -<\/span> dat$<\/span>method2\n><\/span> dat$<\/span>mean <-<\/span> (<\/span>dat$<\/span>method1+<\/span>dat$<\/span>method2)<\/span>\/<\/span>2<\/span>\n><\/span> lm1 <-<\/span> lm<\/span>(<\/span>difference~<\/span>mean,<\/span>data=<\/span>dat)<\/span>\n><\/span> summary<\/span>(<\/span>lm1)<\/span>\nCall:<\/span>\nlm<\/span>(<\/span>formula =<\/span> difference ~<\/span> mean,<\/span> data =<\/span> dat)<\/span>\nResiduals:<\/span>\nMin      1Q  Median      3Q     Max\n-<\/span>72.201<\/span> -<\/span>21.526<\/span>  -<\/span>9.526<\/span>  14.508<\/span>  76.980<\/span>\nCoefficients:<\/span>\nEstimate Std. Error t value Pr<\/span>(<\/span>>|<\/span>t|<\/span>)<\/span>\n(<\/span>Intercept)<\/span> -<\/span>15.06750<\/span>   40.97628<\/span>  -<\/span>0.368<\/span>    0.718<\/span>\nmean          0.02869<\/span>    0.08821<\/span>   0.325<\/span>    0.749<\/span>\nResidual standard error:<\/span> 39.9<\/span> on 15<\/span> degrees of freedom\nMultiple R-<\/span>squared:<\/span>  0.007002<\/span>,<\/span>  Adjusted R-<\/span>squared:<\/span>  -<\/span>0.0592<\/span>\nF<\/span>-<\/span>statistic:<\/span> 0.1058<\/span> on 1<\/span> and 15<\/span> DF,<\/span>  p-<\/span>value:<\/span> 0.7495<\/span>\n<\/code><\/pre>\n\n\n\n
\u30d6\u30e9\u30f3\u30c9\u30a2\u30eb\u30c8\u30de\u30f3\u5206\u6790\u3092 blandr \u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u4f7f\u308f\u305astep by step\u3067\u3084\u3063\u3066\u307f\u308b<\/h2>\n\n\n\n
\u4ee5\u4e0b\u3092\u8a08\u7b97\u3059\u308b\uff1a<\/p>\n\n\n\n