{"id":4014,"date":"2025-06-17T07:18:10","date_gmt":"2025-06-16T22:18:10","guid":{"rendered":"https:\/\/best-biostatistics.com\/toukei-er\/?p=4014"},"modified":"2025-06-22T21:36:37","modified_gmt":"2025-06-22T12:36:37","slug":"generalized-linear-mixed-models-glmm-fundamentals-and-applications","status":"publish","type":"post","link":"https:\/\/best-biostatistics.com\/toukei-er\/entry\/generalized-linear-mixed-models-glmm-fundamentals-and-applications\/","title":{"rendered":"\u4e00\u822c\u5316\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb(GLMM)\u306e\u57fa\u672c\u3068\u5fdc\u7528"},"content":{"rendered":"\n<p>\u81e8\u5e8a\u7814\u7a76\u3084\u751f\u7269\u7d71\u8a08\u5b66\u306e\u5206\u91ce\u3067\u306f\u3001\u60a3\u8005\u3054\u3068\u306e\u3070\u3089\u3064\u304d\u3084\u6e2c\u5b9a\u306e\u53cd\u5fa9\u6027\u306a\u3069\u3001\u30c7\u30fc\u30bf\u304c\u6301\u3064\u8907\u96d1\u306a\u69cb\u9020\u3092\u8003\u616e\u3059\u308b\u3053\u3068\u304c\u4e0d\u53ef\u6b20\u3067\u3042\u308b\u3002\u3057\u304b\u3057\u3001\u57fa\u790e\u7684\u306a\u7d71\u8a08\u30e2\u30c7\u30eb\u3067\u306f\u3001\u3053\u306e\u3088\u3046\u306a\u8907\u96d1\u6027\u3092\u5341\u5206\u306b\u6349\u3048\u304d\u308c\u306a\u3044\u3002\u305d\u3053\u3067\u5fc5\u8981\u3068\u306a\u3063\u3066\u304f\u308b\u306e\u304c\u3001\u300c\u4e00\u822c\u5316\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\uff08Generalized Linear Mixed Model: GLMM\uff09\u300d\u3067\u3042\u308b\u3002GLMM\u306f\u3001\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\u306e\u67d4\u8edf\u6027\u3068\u4e00\u822c\u5316\u7dda\u5f62\u30e2\u30c7\u30eb\u306e\u591a\u69d8\u6027\u3092\u7d44\u307f\u5408\u308f\u305b\u308b\u3053\u3068\u3067\u3001\u3055\u307e\u3056\u307e\u306a\u30bf\u30a4\u30d7\u306e\u5fdc\u7b54\u5909\u6570\u3068\u8907\u96d1\u306a\u30c7\u30fc\u30bf\u69cb\u9020\u3092\u540c\u6642\u306b\u5206\u6790\u3067\u304d\u308b\u5f37\u529b\u306a\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3067\u3042\u308b\u3002\u672c\u8a18\u4e8b\u3067\u306f\u3001GLMM\u306e\u6982\u8981\u304b\u3089\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\u3068\u306e\u9055\u3044\u3001\u81e8\u5e8a\u7814\u7a76\u306b\u304a\u3051\u308b\u5177\u4f53\u7684\u306a\u5fdc\u7528\u4f8b\u3001R\u3067\u306e\u8a08\u7b97\u4f8b\u307e\u3067\u3001\u308f\u304b\u308a\u3084\u3059\u304f\u89e3\u8aac\u3059\u308b\u3002<\/p>\n\n\n\n<!--more-->\n\n\n\n<h2 class=\"wp-block-heading\">\u4e00\u822c\u5316\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\u306e\u6982\u8981<\/h2>\n\n\n\n<p>\u4e00\u822c\u5316\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\uff08GLMM\uff09\u306f\u3001\u56fa\u5b9a\u52b9\u679c\u3068\u5909\u91cf\u52b9\u679c\u306e\u4e21\u65b9\u3092\u542b\u3080\u7d71\u8a08\u30e2\u30c7\u30eb\u3067\u3042\u308a\u3001\u5fdc\u7b54\u5909\u6570\u304c\u6b63\u898f\u5206\u5e03\u306b\u5f93\u308f\u306a\u3044\u5834\u5408\u3067\u3082\u9069\u7528\u53ef\u80fd\u3067\u3042\u308b\u3002\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\u304c\u6b63\u898f\u5206\u5e03\u306b\u5f93\u3046\u5fdc\u7b54\u5909\u6570\u306b\u9650\u5b9a\u3055\u308c\u308b\u306e\u306b\u5bfe\u3057\u3001GLMM\u306f\u3001\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\uff08\u4e8c\u9805\u30c7\u30fc\u30bf\uff09\u3001\u30dd\u30a2\u30bd\u30f3\u56de\u5e30\uff08\u30ab\u30a6\u30f3\u30c8\u30c7\u30fc\u30bf\uff09\u3001\u9806\u5e8f\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\uff08\u9806\u5e8f\u30ab\u30c6\u30b4\u30ea\u30c7\u30fc\u30bf\uff09\u306a\u3069\u3001\u4e00\u822c\u5316\u7dda\u5f62\u30e2\u30c7\u30eb\u306e\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u304c\u6301\u3064\u69d8\u3005\u306a\u5206\u5e03\u3068\u30ea\u30f3\u30af\u95a2\u6570\u3092\u3001\u5909\u91cf\u52b9\u679c\u306e\u6982\u5ff5\u3068\u7d71\u5408\u3057\u305f\u30e2\u30c7\u30eb\u3067\u3042\u308b\u3002<\/p>\n\n\n\n<p>GLMM\u306e\u4e3b\u306a\u7279\u5fb4\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3067\u3042\u308b\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u975e\u6b63\u898f\u5206\u5e03\u306e\u5fdc\u7b54\u5909\u6570\u306b\u5bfe\u5fdc<\/strong>: \u4e8c\u9805\u3001\u30dd\u30a2\u30bd\u30f3\u3001\u30ac\u30f3\u30de\u3001\u8ca0\u306e\u4e8c\u9805\u5206\u5e03\u306a\u3069\u3001\u69d8\u3005\u306a\u5206\u5e03\u306b\u5bfe\u5fdc\u3002<\/li>\n\n\n\n<li><strong>\u30ea\u30f3\u30af\u95a2\u6570<\/strong>: \u5fdc\u7b54\u5909\u6570\u306e\u671f\u5f85\u5024\u3068\u7dda\u5f62\u4e88\u6e2c\u5b50\u3092\u95a2\u9023\u4ed8\u3051\u308b\u95a2\u6570\uff08\u4f8b\uff1a\u30ed\u30b8\u30c3\u30c8\u3001\u5bfe\u6570\u3001\u9006\u6570\uff09\u3002<\/li>\n\n\n\n<li><strong>\u56fa\u5b9a\u52b9\u679c<\/strong>: \u7814\u7a76\u30c7\u30b6\u30a4\u30f3\u306b\u3088\u3063\u3066\u6c7a\u5b9a\u3055\u308c\u308b\u3001\u30b0\u30eb\u30fc\u30d7\u9593\u306e\u5e73\u5747\u7684\u306a\u52b9\u679c\u3002<\/li>\n\n\n\n<li><strong>\u5909\u91cf\u52b9\u679c<\/strong>: \u500b\u4f53\u5dee\u3084\u65bd\u8a2d\u5dee\u306a\u3069\u3001\u30e9\u30f3\u30c0\u30e0\u306a\u3070\u3089\u3064\u304d\u3092\u8868\u73fe\u3059\u308b\u52b9\u679c\u3002\u3053\u308c\u306b\u3088\u308a\u3001\u540c\u3058\u500b\u4f53\u304b\u3089\u306e\u53cd\u5fa9\u6e2c\u5b9a\u3084\u3001\u30af\u30e9\u30b9\u30bf\u30fc\u5316\u3055\u308c\u305f\u30c7\u30fc\u30bf\u306a\u3069\u3001\u30c7\u30fc\u30bf\u306e\u76f8\u95a2\u69cb\u9020\u3092\u9069\u5207\u306b\u30e2\u30c7\u30eb\u5316\u3067\u304d\u308b\u3002<\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\u3068\u306e\u9055\u3044<\/h3>\n\n\n\n<p>\u4e00\u822c\u5316\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\u3068\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\u306e\u4e3b\u306a\u9055\u3044\u306f\u3001<strong>\u5fdc\u7b54\u5909\u6570\u306e\u5206\u5e03<\/strong>\u3067\u3042\u308b\u3002<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><tbody><tr><td>\u7279\u5fb4<\/td><td>\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb (LMM)<\/td><td>\u4e00\u822c\u5316\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb (GLMM)<\/td><\/tr><tr><td>\u5fdc\u7b54\u5909\u6570\u306e\u5206\u5e03<\/td><td>\u6b63\u898f\u5206\u5e03\u306b\u9650\u5b9a\u3055\u308c\u308b<\/td><td>\u4e8c\u9805\u3001\u30dd\u30a2\u30bd\u30f3\u3001\u30ac\u30f3\u30de\u3001\u8ca0\u306e\u4e8c\u9805\u5206\u5e03\u306a\u3069\u3001\u591a\u69d8\u306a\u5206\u5e03\u306b\u5bfe\u5fdc<\/td><\/tr><tr><td>\u30ea\u30f3\u30af\u95a2\u6570<\/td><td>\u6052\u7b49\u95a2\u6570\uff08Identity link\uff09\u304c\u4e00\u822c\u7684<\/td><td>\u30ed\u30b8\u30c3\u30c8\u3001\u5bfe\u6570\u3001\u9006\u6570\u306a\u3069\u3001\u5fdc\u7b54\u5909\u6570\u306e\u5206\u5e03\u306b\u5fdc\u3058\u305f\u95a2\u6570\u3092\u4f7f\u7528<\/td><\/tr><tr><td>\u4e3b\u306a\u7528\u9014<\/td><td>\u9023\u7d9a\u30c7\u30fc\u30bf\u3067\u3001\u6b8b\u5dee\u304c\u6b63\u898f\u5206\u5e03\u306b\u5f93\u3046\u3068\u4eee\u5b9a\u3067\u304d\u308b\u5834\u5408<\/td><td>\u4e8c\u5024\u3001\u30ab\u30a6\u30f3\u30c8\u3001\u9806\u5e8f\u30c7\u30fc\u30bf\u306a\u3069\u3001\u975e\u6b63\u898f\u5206\u5e03\u306b\u5f93\u3046\u30c7\u30fc\u30bf<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p>LMM\u306f\u3001\u4f8b\u3048\u3070\u8840\u5727\u3084\u4f53\u91cd\u3068\u3044\u3063\u305f\u9023\u7d9a\u7684\u306a\u6e2c\u5b9a\u5024\u3067\u3001\u3053\u308c\u3089\u306e\u6e2c\u5b9a\u5024\u306e\u3070\u3089\u3064\u304d\u304c\u6b63\u898f\u5206\u5e03\u306b\u5f93\u3046\u3068\u4eee\u5b9a\u3067\u304d\u308b\u5834\u5408\u306b\u9069\u3057\u3066\u3044\u308b\u3002\u4e00\u65b9\u3001GLMM\u306f\u3001\u75be\u60a3\u306e\u6709\u7121\uff08\u4e8c\u5024\uff09\u3001\u30a4\u30d9\u30f3\u30c8\u306e\u767a\u751f\u56de\u6570\uff08\u30ab\u30a6\u30f3\u30c8\uff09\u3001\u75bc\u75db\u306e\u7a0b\u5ea6\uff08\u9806\u5e8f\uff09\u306a\u3069\u3001\u6b63\u898f\u5206\u5e03\u3092\u4eee\u5b9a\u3067\u304d\u306a\u3044\u591a\u69d8\u306a\u30bf\u30a4\u30d7\u306e\u30c7\u30fc\u30bf\u306b\u5bfe\u5fdc\u3067\u304d\u308b\u305f\u3081\u3001\u3088\u308a\u5e83\u7bc4\u306a\u81e8\u5e8a\u7814\u7a76\u306b\u5fdc\u7528\u53ef\u80fd\u3067\u3042\u308b\u3002<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\u81e8\u5e8a\u7814\u7a76\u306e\u5177\u4f53\u4f8b<\/h2>\n\n\n\n<p>GLMM\u306f\u3001\u8907\u96d1\u306a\u81e8\u5e8a\u30c7\u30fc\u30bf\u3092\u5206\u6790\u3059\u308b\u305f\u3081\u306e\u5f37\u529b\u306a\u30c4\u30fc\u30eb\u3067\u3042\u308b\u3002\u4ee5\u4e0b\u306b\u5177\u4f53\u7684\u306a\u4f8b\u3092\u6319\u3052\u308b\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">\u4e8c\u9805\u30a4\u30d9\u30f3\u30c8\u306e\u4f8b\uff1a\u65b0\u85ac\u306e\u6295\u4e0e\u306b\u3088\u308b\u75be\u60a3\u306e\u518d\u767a\u6709\u7121<\/h3>\n\n\n\n<p>\u3042\u308b\u75be\u60a3\u306e\u60a3\u8005\u3092\u5bfe\u8c61\u306b\u3001\u65b0\u85ac\u3068\u30d7\u30e9\u30bb\u30dc\u306e\u6709\u52b9\u6027\u3092\u6bd4\u8f03\u3059\u308b\u81e8\u5e8a\u8a66\u9a13\u3092\u8003\u3048\u308b\u3002\u5404\u60a3\u8005\u306f\u8907\u6570\u56de\u306e\u8a3a\u5bdf\u3092\u53d7\u3051\u3001\u305d\u308c\u305e\u308c\u306e\u8a3a\u5bdf\u6642\u306b\u75be\u60a3\u306e\u518d\u767a\u6709\u7121\uff08\u518d\u767a\u3042\u308a\/\u306a\u3057\uff09\u304c\u8a18\u9332\u3055\u308c\u308b\u3002\u3053\u306e\u30c7\u30fc\u30bf\u3067\u306f\u3001\u540c\u3058\u60a3\u8005\u304b\u3089\u306e\u518d\u767a\u6709\u7121\u306e\u6e2c\u5b9a\u306f\u72ec\u7acb\u3067\u306f\u306a\u3044\uff08\u6642\u9593\u7684\u306a\u76f8\u95a2\u304c\u3042\u308b\u53ef\u80fd\u6027\u304c\u9ad8\u3044\uff09\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u5fdc\u7b54\u5909\u6570<\/strong>: \u75be\u60a3\u306e\u518d\u767a\u6709\u7121\uff08\u4e8c\u5024\u30c7\u30fc\u30bf\uff1a0=\u306a\u3057, 1=\u3042\u308a\uff09<\/li>\n\n\n\n<li><strong>\u56fa\u5b9a\u52b9\u679c<\/strong>: \u6cbb\u7642\u7fa4\uff08\u65b0\u85ac vs \u30d7\u30e9\u30bb\u30dc\uff09\u3001\u8a3a\u5bdf\u6642\u671f<\/li>\n\n\n\n<li><strong>\u5909\u91cf\u52b9\u679c<\/strong>: \u5404\u60a3\u8005\u306e\u30d9\u30fc\u30b9\u30e9\u30a4\u30f3\u306e\u518d\u767a\u50be\u5411\u306e\u3070\u3089\u3064\u304d\uff08\u60a3\u8005\u3054\u3068\u306e\u5207\u7247\uff09<\/li>\n<\/ul>\n\n\n\n<p>\u3053\u306e\u5834\u5408\u3001\u30ed\u30b8\u30c3\u30c8\u30ea\u30f3\u30af\u95a2\u6570\u3068\u4e8c\u9805\u5206\u5e03\u3092\u4eee\u5b9a\u3057\u305fGLMM\uff08\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u6df7\u5408\u30e2\u30c7\u30eb\uff09\u3092\u7528\u3044\u308b\u3053\u3068\u3067\u3001\u6cbb\u7642\u7fa4\u3068\u8a3a\u5bdf\u6642\u671f\u304c\u518d\u767a\u306b\u4e0e\u3048\u308b\u5f71\u97ff\u3092\u8a55\u4fa1\u3057\u3064\u3064\u3001\u60a3\u8005\u3054\u3068\u306e\u3070\u3089\u3064\u304d\u3092\u9069\u5207\u306b\u8003\u616e\u3057\u305f\u89e3\u6790\u304c\u53ef\u80fd\u3067\u3042\u308b\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">\u9806\u5e8f\u30ab\u30c6\u30b4\u30ea\u306e\u4f8b\uff1a\u85ac\u5264\u6295\u4e0e\u306b\u3088\u308b\u75c7\u72b6\u6539\u5584\u5ea6\u306e\u8a55\u4fa1<\/h3>\n\n\n\n<p>\u95a2\u7bc0\u708e\u60a3\u8005\u3092\u5bfe\u8c61\u306b\u3001\u65b0\u85ac\u304c\u75c7\u72b6\u6539\u5584\u306b\u4e0e\u3048\u308b\u5f71\u97ff\u3092\u8a55\u4fa1\u3059\u308b\u8a66\u9a13\u3092\u8003\u3048\u308b\u3002\u60a3\u8005\u306f\u4e00\u5b9a\u671f\u9593\u3001\u85ac\u5264\u3092\u670d\u7528\u3057\u3001\u5b9a\u671f\u7684\u306b\u75c7\u72b6\u306e\u6539\u5584\u5ea6\u3092\u300c\u5168\u304f\u6539\u5584\u306a\u3057\u300d\u300c\u5c11\u3057\u6539\u5584\u300d\u300c\u304b\u306a\u308a\u6539\u5584\u300d\u300c\u5b8c\u5168\u306b\u6539\u5584\u300d\u306e4\u6bb5\u968e\u3067\u8a55\u4fa1\u3059\u308b\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u5fdc\u7b54\u5909\u6570<\/strong>: \u75c7\u72b6\u6539\u5584\u5ea6\uff08\u9806\u5e8f\u30ab\u30c6\u30b4\u30ea\u30c7\u30fc\u30bf\uff1a1=\u5168\u304f\u6539\u5584\u306a\u3057, 2=\u5c11\u3057\u6539\u5584, 3=\u304b\u306a\u308a\u6539\u5584, 4=\u5b8c\u5168\u306b\u6539\u5584\uff09<\/li>\n\n\n\n<li><strong>\u56fa\u5b9a\u52b9\u679c<\/strong>: \u6cbb\u7642\u7fa4\u3001\u6295\u4e0e\u671f\u9593<\/li>\n\n\n\n<li><strong>\u5909\u91cf\u52b9\u679c<\/strong>: \u5404\u60a3\u8005\u306e\u75c7\u72b6\u6539\u5584\u50be\u5411\u306e\u3070\u3089\u3064\u304d\uff08\u60a3\u8005\u3054\u3068\u306e\u5207\u7247\uff09<\/li>\n<\/ul>\n\n\n\n<p>\u3053\u306e\u5834\u5408\u3001\u7d2f\u7a4d\u30ed\u30b8\u30c3\u30c8\u30ea\u30f3\u30af\u95a2\u6570\u3068\u591a\u9805\u5206\u5e03\uff08\u307e\u305f\u306f\u9806\u5e8f\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\uff09\u3092\u4eee\u5b9a\u3057\u305fGLMM\uff08\u9806\u5e8f\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u6df7\u5408\u30e2\u30c7\u30eb\uff09\u3092\u7528\u3044\u308b\u3053\u3068\u3067\u3001\u85ac\u5264\u304c\u75c7\u72b6\u6539\u5584\u5ea6\u306b\u4e0e\u3048\u308b\u5f71\u97ff\u3092\u8a55\u4fa1\u3057\u3001\u60a3\u8005\u9593\u306e\u3070\u3089\u3064\u304d\u3092\u8003\u616e\u3057\u305f\u89e3\u6790\u304c\u53ef\u80fd\u3067\u3042\u308b\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">\u30ab\u30a6\u30f3\u30c8\u30c7\u30fc\u30bf\u306e\u4f8b\uff1a\u5598\u606f\u60a3\u8005\u306e\u5e74\u9593\u5165\u9662\u56de\u6570<\/h3>\n\n\n\n<p>\u5598\u606f\u60a3\u8005\u3092\u5bfe\u8c61\u306b\u3001\u65b0\u3057\u3044\u6cbb\u7642\u6cd5\u306e\u52b9\u679c\u3092\u8a55\u4fa1\u3059\u308b\u7814\u7a76\u3092\u8003\u3048\u308b\u3002\u5404\u60a3\u8005\u306b\u3064\u3044\u3066\u3001\u6cbb\u7642\u958b\u59cb\u5f8c\u306e\u5e74\u9593\u5165\u9662\u56de\u6570\u304c\u30a2\u30a6\u30c8\u30ab\u30e0\u3067\u3042\u308b\u3002\u5165\u9662\u56de\u6570\u306f\u975e\u8ca0\u306e\u6574\u6570\u3067\u3042\u308a\u3001\u3057\u3070\u3057\u3070\u53f3\u306b\u6b6a\u3093\u3060\uff08\u88fe\u3092\u5f15\u3044\u305f\uff09\u5206\u5e03\u3092\u793a\u3059\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u5fdc\u7b54\u5909\u6570<\/strong>: \u5e74\u9593\u5165\u9662\u56de\u6570\uff08\u30ab\u30a6\u30f3\u30c8\u30c7\u30fc\u30bf\uff09<\/li>\n\n\n\n<li><strong>\u56fa\u5b9a\u52b9\u679c<\/strong>: \u6cbb\u7642\u6cd5\uff08\u65b0\u6cbb\u7642 vs \u6a19\u6e96\u6cbb\u7642\uff09\u3001\u5e74\u9f62\u3001\u55ab\u7159\u6b74<\/li>\n\n\n\n<li><strong>\u5909\u91cf\u52b9\u679c<\/strong>: \u5404\u60a3\u8005\u306e\u30d9\u30fc\u30b9\u30e9\u30a4\u30f3\u306e\u5165\u9662\u30ea\u30b9\u30af\u306e\u3070\u3089\u3064\u304d\uff08\u60a3\u8005\u3054\u3068\u306e\u5207\u7247\uff09<\/li>\n<\/ul>\n\n\n\n<p>\u3053\u306e\u5834\u5408\u3001\u5bfe\u6570\u30ea\u30f3\u30af\u95a2\u6570\u3068\u30dd\u30a2\u30bd\u30f3\u5206\u5e03\u3092\u4eee\u5b9a\u3057\u305fGLMM\uff08\u30dd\u30a2\u30bd\u30f3\u6df7\u5408\u30e2\u30c7\u30eb\uff09\u3092\u7528\u3044\u308b\u3053\u3068\u3067\u3001\u6cbb\u7642\u6cd5\u304c\u5e74\u9593\u5165\u9662\u56de\u6570\u306b\u4e0e\u3048\u308b\u5f71\u97ff\u3092\u8a55\u4fa1\u3057\u3001\u60a3\u8005\u3054\u3068\u306e\u5165\u9662\u56de\u6570\u306e\u3070\u3089\u3064\u304d\u3092\u9069\u5207\u306b\u8003\u616e\u3057\u305f\u89e3\u6790\u304c\u53ef\u80fd\u3067\u3042\u308b\u3002\u3082\u3057\u30ab\u30a6\u30f3\u30c8\u30c7\u30fc\u30bf\u306b\u904e\u5206\u6563\u304c\u3042\u308b\u5834\u5408\uff08\u5206\u6563\u304c\u5e73\u5747\u3088\u308a\u5927\u304d\u3044\u5834\u5408\uff09\u306f\u3001\u8ca0\u306e\u4e8c\u9805\u5206\u5e03\u3092\u4eee\u5b9a\u3057\u305fGLMM\u3092\u7528\u3044\u308b\u3053\u3068\u3082\u6709\u52b9\u3067\u3042\u308b\u3002<\/p>\n\n\n\n<div id=\"biost-1381209431\" class=\"biost- biost-entity-placement\"><p style=\"text-align: center;\"><span style=\"font-size: 20px;\"><strong><a href=\"https:\/\/best-biostatistics.com\/kmhl\">\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 href=\"https:\/\/best-biostatistics.com\/kmhl\"><img class=\"aligncenter wp-image-2794 size-full\" src=\"https:\/\/best-biostatistics.com\/wp\/wp-content\/uploads\/2023\/11\/bn_r_03.png\" alt=\"\" width=\"500\" height=\"327\" \/><\/a>\r\n<p style=\"text-align: center;\"><span style=\"color: #ff0000; font-size: 20px;\"><strong><span class=\"marker2\">\u21911\u4e07\u4eba\u4ee5\u4e0a\u306e\u533b\u7642\u5f93\u4e8b\u8005\u304c\u8cfc\u8aad\u4e2d<\/span><\/strong><\/span><\/p><\/div><h2 class=\"wp-block-heading\">R \u3067\u306e\u8a08\u7b97\u4f8b<\/h2>\n\n\n\n<p>R \u3067\u306f\u3001<code>lme4<\/code>\u30d1\u30c3\u30b1\u30fc\u30b8\u3084<code>glmmTMB<\/code>\u30d1\u30c3\u30b1\u30fc\u30b8\u306a\u3069\u3092\u7528\u3044\u3066GLMM\u3092\u7c21\u5358\u306b\u5b9f\u88c5\u3067\u304d\u308b\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>\u4e8c\u9805\u30a4\u30d9\u30f3\u30c8\u306e\u8a08\u7b97\u4f8b<\/strong><\/h3>\n\n\n\n<p>\u3053\u3053\u3067\u306f\u3001<code>lme4<\/code>\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u4f7f\u3063\u305f\u4e8c\u9805\u30a4\u30d9\u30f3\u30c8\u306e\u7c21\u5358\u306a\u4f8b\u3092\u793a\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># \u5fc5\u8981\u306b\u5fdc\u3058\u3066\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\n# install.packages(\"lme4\")\n# install.packages(\"dplyr\") # \u30c7\u30fc\u30bf\u64cd\u4f5c\u7528\n# install.packages(\"ggplot2\") # \u53ef\u8996\u5316\u7528\n\nlibrary(lme4)\nlibrary(dplyr)\nlibrary(ggplot2)\n\n# \u4f8b\u3068\u3057\u3066\u3001\u67b6\u7a7a\u306e\u30c7\u30fc\u30bf\u3092\u751f\u6210\n# 20\u4eba\u306e\u60a3\u8005\u304c3\u56de\u305a\u3064\u8a3a\u5bdf\u3092\u53d7\u3051\u3001\u518d\u767a\u306e\u6709\u7121\u3092\u8a18\u9332\nset.seed(123)\nn_patients &lt;- 20\nn_visits &lt;- 3\n\npatient_id &lt;- rep(1:n_patients, each = n_visits)\nvisit &lt;- rep(1:n_visits, times = n_patients)\ntreatment &lt;- sample(c(\"NewDrug\", \"Placebo\"), n_patients, replace = TRUE) %>% rep(each = n_visits)\n\n# \u518d\u767a\u306e\u78ba\u7387\u3092\u6cbb\u7642\u7fa4\u3068\u8a3a\u5bdf\u6642\u671f\u306b\u5fdc\u3058\u3066\u8a2d\u5b9a\n# \u65b0\u85ac\u306e\u65b9\u304c\u518d\u767a\u78ba\u7387\u304c\u4f4e\u3044\u3001\u8a3a\u5bdf\u56de\u6570\u304c\u5897\u3048\u308b\u3054\u3068\u306b\u518d\u767a\u78ba\u7387\u304c\u5909\u5316\ntrue_prob &lt;- case_when(\n  treatment == \"NewDrug\" &amp; visit == 1 ~ 0.2,\n  treatment == \"NewDrug\" &amp; visit == 2 ~ 0.15,\n  treatment == \"NewDrug\" &amp; visit == 3 ~ 0.1,\n  treatment == \"Placebo\" &amp; visit == 1 ~ 0.4,\n  treatment == \"Placebo\" &amp; visit == 2 ~ 0.35,\n  treatment == \"Placebo\" &amp; visit == 3 ~ 0.3,\n  TRUE ~ NA_real_ # \u3042\u308a\u3048\u306a\u3044\u5024\u306fNA\n)\n\n# \u60a3\u8005\u3054\u3068\u306e\u30e9\u30f3\u30c0\u30e0\u306a\u52b9\u679c\uff08\u5207\u7247\uff09\u3092\u8ffd\u52a0\nrandom_effect_patient &lt;- rnorm(n_patients, mean = 0, sd = 0.5) %>% rep(each = n_visits)\n\n# \u30ed\u30b8\u30c3\u30c8\u5909\u63db\u3057\u305f\u771f\u306e\u78ba\u7387\u306b\u30e9\u30f3\u30c0\u30e0\u52b9\u679c\u3092\u52a0\u3048\u308b\nlogit_prob_with_re &lt;- log(true_prob \/ (1 - true_prob)) + random_effect_patient\n\n# \u9006\u30ed\u30b8\u30c3\u30c8\u5909\u63db\u3057\u3066\u3001\u500b\u5225\u306e\u89b3\u5bdf\u306b\u304a\u3051\u308b\u78ba\u7387\u3092\u8a08\u7b97\nprob_outcome &lt;- exp(logit_prob_with_re) \/ (1 + exp(logit_prob_with_re))\n\n# \u7d50\u679c\uff08\u518d\u767a\u6709\u7121\uff09\u3092\u751f\u6210\nrecurrence &lt;- rbinom(n_patients * n_visits, 1, prob_outcome)\n\n# \u30c7\u30fc\u30bf\u30d5\u30ec\u30fc\u30e0\u306e\u4f5c\u6210\ndf_glmm &lt;- data.frame(\n  patient_id = factor(patient_id),\n  visit = factor(visit),\n  treatment = factor(treatment),\n  recurrence = recurrence\n)\n\n# \u30c7\u30fc\u30bf\u306e\u78ba\u8a8d\nhead(df_glmm)\nstr(df_glmm)\n\n# GLMM\u306e\u5b9f\u884c\uff08\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u6df7\u5408\u30e2\u30c7\u30eb\uff09\n# recurrence ~ treatment + visit: \u56fa\u5b9a\u52b9\u679c\n# (1 | patient_id): patient_id\u3054\u3068\u306e\u5207\u7247\u3092\u5909\u91cf\u52b9\u679c\u3068\u3057\u3066\u8003\u616e\nmodel_glmm &lt;- glmer(recurrence ~ treatment + visit + (1 | patient_id), \n                    data = df_glmm, \n                    family = binomial(link = \"logit\"))\n\n# \u30e2\u30c7\u30eb\u7d50\u679c\u306e\u8868\u793a\nsummary(model_glmm)\n\n# \u30aa\u30c3\u30ba\u6bd4\u306e\u8a08\u7b97\uff08\u89e3\u91c8\u3092\u5bb9\u6613\u306b\u3059\u308b\u305f\u3081\uff09\nexp(fixef(model_glmm))\n\n# \u56fa\u5b9a\u52b9\u679c\u306e95%\u4fe1\u983c\u533a\u9593\u3092\u8a08\u7b97\nconfint_fixed &lt;- confint(model_glmm, method=\"Wald\", oldNames=FALSE)&#91;-1,]\nprint(\"\u56fa\u5b9a\u52b9\u679c\u306e95%\u4fe1\u983c\u533a\u9593:\")\nprint(confint_fixed)\n\n# \u30aa\u30c3\u30ba\u6bd4\u3068\u305d\u306e95%\u4fe1\u983c\u533a\u9593\nodds_ratios &lt;- data.frame(\n  OR = round(exp(fixef(model_glmm))&#91;-1], 2),\n  CI_lower = round(exp(confint_fixed&#91;-1,1]), 2),\n  CI_upper = round(exp(confint_fixed&#91;-1,2]), 2),\n  p_value = round(summary(model_glmm)$coefficients&#91;-1,4], 3)\n)\nprint(\"\u30aa\u30c3\u30ba\u6bd4\u3068\u305d\u306e95%\u4fe1\u983c\u533a\u9593:\")\nprint(odds_ratios)<\/code><\/pre>\n\n\n\n<p><strong>\u5b9f\u884c\u7d50\u679c\uff1a<\/strong><\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>> # \u30e2\u30c7\u30eb\u7d50\u679c\u306e\u8868\u793a\n> summary(model_glmm)\nGeneralized linear mixed model fit by maximum likelihood (Laplace\n  Approximation) &#91;glmerMod]\n Family: binomial  ( logit )\nFormula: recurrence ~ treatment + visit + (1 | patient_id)\n   Data: df_glmm\n\n      AIC       BIC    logLik -2*log(L)  df.resid \n     71.7      82.2     -30.8      61.7        55 \n\nScaled residuals: \n    Min      1Q  Median      3Q     Max \n-0.8745 -0.5908 -0.4233 -0.3546  2.8200 \n\nRandom effects:\n Groups     Name        Variance Std.Dev.\n patient_id (Intercept) 0        0       \nNumber of obs: 60, groups:  patient_id, 20\n\nFixed effects:\n                 Estimate Std. Error z value Pr(>|z|)\n(Intercept)       -0.9354     0.5696  -1.642    0.101\ntreatmentPlacebo   0.6671     0.6318   1.056    0.291\nvisit2            -0.7841     0.7379  -1.063    0.288\nvisit3            -1.1381     0.7906  -1.440    0.150\n\nCorrelation of Fixed Effects:\n            (Intr) trtmnP visit2\ntretmntPlcb -0.553              \nvisit2      -0.512 -0.044       \nvisit3      -0.471 -0.054  0.389\noptimizer (Nelder_Mead) convergence code: 0 (OK)\nboundary (singular) fit: see help('isSingular')<\/code><\/pre>\n\n\n\n<p>treatmentPlacebo \u306e estimate \u304c\u30d7\u30e9\u30bb\u30dc\u7fa4\u306e\u518d\u767a\u30aa\u30c3\u30ba\u304c\u65b0\u85ac\u7fa4\u3068\u6bd4\u8f03\u3057\u3066\u3069\u308c\u304f\u3089\u3044\u9ad8\u3044\u304b\u3092\u8868\u3057\u3066\u3044\u308b\u3002\u305f\u3060\u3057\u3001\u5bfe\u6570\u30aa\u30c3\u30ba\u6bd4\u306b\u306a\u3063\u3066\u3044\u308b\u306e\u3067\u3001\u306e\u3061\u307b\u3069\u3001\u771f\u6570\u306b\u5909\u63db\u3057\u3066\u3001\u78ba\u8a8d\u3059\u308b\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>&gt; print(\"\u30aa\u30c3\u30ba\u6bd4\u3068\u305d\u306e95%\u4fe1\u983c\u533a\u9593:\")\n&#91;1] \"\u30aa\u30c3\u30ba\u6bd4\u3068\u305d\u306e95%\u4fe1\u983c\u533a\u9593:\"\n&gt; print(odds_ratios)\n                   OR CI_lower CI_upper p_value\ntreatmentPlacebo 1.95     0.56     6.72   0.291\nvisit2           0.46     0.11     1.94   0.288\nvisit3           0.32     0.07     1.51   0.150<\/code><\/pre>\n\n\n\n<p>\u3053\u306e\u7d50\u679c\u304c\u3001\u771f\u6570\u306b\u3057\u305f\u7d50\u679c\u3067\u3042\u308b\u3002treatmentPlacebo \u306e 1.95 \u304c\u30d7\u30e9\u30bb\u30dc\u7fa4\u306e\u518d\u767a\u30aa\u30c3\u30ba\u6bd4\u306e\u63a8\u5b9a\u5024\u3067\u3042\u308b\u3002\u305f\u3060\u3057\u300195\uff05\u4fe1\u983c\u533a\u9593\u3092\u898b\u308b\u3068\u30011\u3088\u308a\u5c0f\u3055\u3044\uff08\u3064\u307e\u308a\u3001\u95a2\u9023\u3057\u306a\u3044\u65b9\u5411\u306e\uff09\u53ef\u80fd\u6027\u3082\u3042\u308b\u305f\u3081\u3001\u540d\u76ee\u4e0a\u3067\u3042\u3063\u3066\u3082\u3001\u7d71\u8a08\u5b66\u7684\u6709\u610f\u306a\u7d50\u679c\u3067\u306f\u306a\u3044\u3068\u304d\u306f\u3001\u7d50\u679c\u306e\u89e3\u91c8\u30fb\u4f4d\u7f6e\u3065\u3051\u3092\u614e\u91cd\u306b\u3057\u306a\u3044\u3068\u3044\u3051\u306a\u3044\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">\u9806\u5e8f\u30ab\u30c6\u30b4\u30ea\u306e\u8a08\u7b97\u4f8b<\/h3>\n\n\n\n<p>\u9806\u5e8f\u30ab\u30c6\u30b4\u30ea\u30c7\u30fc\u30bf\u3092\u4e00\u822c\u5316\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\u3067\u89e3\u6790\u3059\u308b\u5834\u5408\u3001<code>ordinal<\/code>\u30d1\u30c3\u30b1\u30fc\u30b8\u306e<code>clmm()<\/code>\u95a2\u6570\u304c\u4fbf\u5229\u3067\u3042\u308b\u3002\u3053\u306e\u95a2\u6570\u306f\u3001\u7d2f\u7a4d\u30ed\u30b8\u30c3\u30c8\u30ea\u30f3\u30af\u95a2\u6570\u3092\u7528\u3044\u305f\u9806\u5e8f\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u6df7\u5408\u30e2\u30c7\u30eb\u3092\u5b9f\u884c\u3067\u304d\u308b\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># \u5fc5\u8981\u306b\u5fdc\u3058\u3066\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\n# install.packages(\"ordinal\")\n# install.packages(\"dplyr\") # \u30c7\u30fc\u30bf\u64cd\u4f5c\u7528\n\nlibrary(ordinal)\nlibrary(dplyr)\n\n# \u4f8b\u3068\u3057\u3066\u3001\u67b6\u7a7a\u306e\u30c7\u30fc\u30bf\u3092\u751f\u6210\nset.seed(456)\nn_patients &lt;- 60 \nn_visits &lt;- 3\n\npatient_id &lt;- rep(1:n_patients, each = n_visits)\ntreatment &lt;- sample(c(\"DrugA\", \"DrugB\"), n_patients, replace = TRUE) %&gt;% rep(each = n_visits)\nvisit &lt;- rep(c(\"Baseline\", \"Post_Treatment_1\", \"Post_Treatment_2\"), times = n_patients)\n\n# \u60a3\u8005\u3054\u3068\u306e\u30e9\u30f3\u30c0\u30e0\u306a\u52b9\u679c\uff08\u5207\u7247\uff09\u3092\u751f\u6210\nrandom_effect_patient &lt;- rnorm(n_patients, mean = 0, sd = 1.2) %&gt;% rep(each = n_visits)\n\n# \u75c7\u72b6\u6539\u5584\u5ea6\u306e\u771f\u306e\u78ba\u7387\u7684\u306a\u50be\u5411\u3092\u8a2d\u5b9a\nlinear_predictor &lt;- case_when(\n    treatment == \"DrugA\" &amp; visit == \"Baseline\" ~ 0.2,\n    treatment == \"DrugA\" &amp; visit == \"Post_Treatment_1\" ~ 0.8,\n    treatment == \"DrugA\" &amp; visit == \"Post_Treatment_2\" ~ 1.2,\n    treatment == \"DrugB\" &amp; visit == \"Baseline\" ~ 0.1,\n    treatment == \"DrugB\" &amp; visit == \"Post_Treatment_1\" ~ 0.4,\n    treatment == \"DrugB\" &amp; visit == \"Post_Treatment_2\" ~ 0.7,\n    TRUE ~ NA_real_\n)\n\n# \u30e9\u30f3\u30c0\u30e0\u52b9\u679c\u3092\u7dda\u5f62\u4e88\u6e2c\u5b50\u306b\u52a0\u3048\u308b\nlinear_predictor_with_re &lt;- linear_predictor + random_effect_patient\n\n# \u7d2f\u7a4d\u78ba\u7387\u304b\u3089\u75c7\u72b6\u6539\u5584\u5ea6\u3092\u751f\u6210\nthresholds &lt;- c(-1.5, 0.5, 2.0) \n\n# \u75c7\u72b6\u30ab\u30c6\u30b4\u30ea\u3092\u5272\u308a\u5f53\u3066\u308b\u95a2\u6570\nget_ordinal_category &lt;- function(lp_val, thresholds) {\n    prob_cum1 &lt;- plogis(thresholds&#91;1] - lp_val) # P(Y &lt;= 1)\n    prob_cum2 &lt;- plogis(thresholds&#91;2] - lp_val) # P(Y &lt;= 2)\n    prob_cum3 &lt;- plogis(thresholds&#91;3] - lp_val) # P(Y &lt;= 3)\n\n    u &lt;- runif(1) # \u4e00\u69d8\u4e71\u6570\u3092\u751f\u6210\n\n    if (u &lt; prob_cum1) {\n        return(1) # \u5168\u304f\u6539\u5584\u306a\u3057\n    } else if (u &lt; prob_cum2) {\n        return(2) # \u5c11\u3057\u6539\u5584\n    } else if (u &lt; prob_cum3) {\n        return(3) # \u304b\u306a\u308a\u6539\u5584\n    } else {\n        return(4) # \u5b8c\u5168\u306b\u6539\u5584\n    }\n}\n\n# \u5404\u884c\u306b\u3064\u3044\u3066\u75c7\u72b6\u6539\u5584\u5ea6\u3092\u751f\u6210\nsymptom_improvement &lt;- mapply(get_ordinal_category, linear_predictor_with_re, MoreArgs = list(thresholds))\n\n# \u30c7\u30fc\u30bf\u30d5\u30ec\u30fc\u30e0\u306e\u4f5c\u6210\ndf_ordinal_glmm &lt;- data.frame(\n    patient_id = factor(patient_id),\n    treatment = factor(treatment),\n    visit = factor(visit, levels = c(\"Baseline\", \"Post_Treatment_1\", \"Post_Treatment_2\")), # \u9806\u5e8f\u3092\u8003\u616e\n    symptom_improvement = ordered(symptom_improvement,\n        levels = 1:4\n    )\n)\n\n# \u30c7\u30fc\u30bf\u306e\u78ba\u8a8d\nhead(df_ordinal_glmm)\nstr(df_ordinal_glmm)\n\n# \u9806\u5e8f\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u6df7\u5408\u30e2\u30c7\u30eb\u306e\u5b9f\u884c\ncat(\"\\n=== \u9806\u5e8f\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u6df7\u5408\u30e2\u30c7\u30eb\u306e\u5b9f\u884c ===\\n\")\nmodel_ordinal_glmm &lt;- clmm(symptom_improvement ~ treatment + visit + (1 | patient_id),\n    data = df_ordinal_glmm)\n\n# \u30e2\u30c7\u30eb\u7d50\u679c\u306e\u8868\u793a\ncat(\"\\n=== \u30e2\u30c7\u30eb\u7d50\u679c ===\\n\")\nprint(summary(model_ordinal_glmm))\n\n# \u30aa\u30c3\u30ba\u6bd4\u306e\u8a08\u7b97\uff08\u89e3\u91c8\u3092\u5bb9\u6613\u306b\u3059\u308b\u305f\u3081\uff09\nexp(coef(model_ordinal_glmm))\n\n# \u56fa\u5b9a\u52b9\u679c\u306e95%\u4fe1\u983c\u533a\u9593\u3092\u8a08\u7b97\nconfint_fixed &lt;- confint(model_ordinal_glmm, method = \"Wald\", oldNames = FALSE)&#91;-1, ]\nprint(\"\u56fa\u5b9a\u52b9\u679c\u306e95%\u4fe1\u983c\u533a\u9593:\")\nprint(exp(confint_fixed))\n\n# \u30aa\u30c3\u30ba\u6bd4\u3068\u305d\u306e95%\u4fe1\u983c\u533a\u9593\nodds_ratios &lt;- data.frame(\n    OR = round(exp(coef(model_ordinal_glmm))&#91;c(-1,-2,-3)], 2),\n    CI_lower = round(exp(confint_fixed&#91;c(-1,-2), 1]), 2),\n    CI_upper = round(exp(confint_fixed&#91;c(-1,-2), 2]), 2),\n    p_value = round(summary(model_ordinal_glmm)$coefficients&#91;c(-1,-2,-3), 4], 3)\n)\nprint(\"\u30aa\u30c3\u30ba\u6bd4\u3068\u305d\u306e95%\u4fe1\u983c\u533a\u9593:\")\nprint(odds_ratios)<\/code><\/pre>\n\n\n\n<p><strong>\u5b9f\u884c\u7d50\u679c\uff1a<\/strong><\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>=== \u30e2\u30c7\u30eb\u7d50\u679c ===\n&gt; print(summary(model_ordinal_glmm))\nCumulative Link Mixed Model fitted with the Laplace approximation\n\nformula: symptom_improvement ~ treatment + visit + (1 | patient_id)\ndata:    df_ordinal_glmm\n\n link  threshold nobs logLik  AIC    niter     max.grad cond.H \n logit flexible  180  -241.99 497.98 342(1028) 3.59e-05 3.1e+01\n\nRandom effects:\n Groups     Name        Variance Std.Dev.\n patient_id (Intercept) 0.5177   0.7195  \nNumber of groups:  patient_id 60 \n\nCoefficients:\n                      Estimate Std. Error z value Pr(&gt;|z|)\ntreatmentDrugB        -0.02135    0.33177  -0.064    0.949\nvisitPost_Treatment_1  0.40715    0.33501   1.215    0.224\nvisitPost_Treatment_2  0.40299    0.33866   1.190    0.234\n\nThreshold coefficients:\n    Estimate Std. Error z value\n1|2  -1.4436     0.3464  -4.167\n2|3   0.3255     0.3174   1.026\n3|4   1.6851     0.3532   4.771<\/code><\/pre>\n\n\n\n<p>treatmentDrugA \u306b\u5bfe\u3059\u308b DrugB \u306e\u5bfe\u6570\u30aa\u30c3\u30ba\u6bd4\u3084 visitBaseline \u306b\u5bfe\u3059\u308b Post_Treatment1, Post_Treatment2 \u306e\u30aa\u30c3\u30ba\u6bd4\u304c\u8a08\u7b97\u3055\u308c\u3066\u3044\u308b\u3002Threshold \u3082\u8a08\u7b97\u3055\u308c\u3066\u3044\u308b\u3002Threshold \u306f\u3001\u6700\u521d\u306e\u60f3\u5b9a\u3068\u6982\u306d\u4e00\u81f4\u3057\u3066\u3044\u308b\u306e\u304c\u308f\u304b\u308b\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>&#91;1] \"\u30aa\u30c3\u30ba\u6bd4\u3068\u305d\u306e95%\u4fe1\u983c\u533a\u9593:\"\n&gt; print(odds_ratios)\n                        OR CI_lower CI_upper p_value\ntreatmentDrugB        0.98     0.51     1.88   0.949\nvisitPost_Treatment_1 1.50     0.78     2.90   0.224\nvisitPost_Treatment_2 1.50     0.77     2.91   0.234<\/code><\/pre>\n\n\n\n<p>\u6700\u5f8c\u306b\u3001Treatment \u3068 Visit \u306b\u95a2\u3059\u308b\u3068\u30aa\u30c3\u30ba\u6bd4\u306895\uff05\u4fe1\u983c\u533a\u9593\u304c\u51fa\u529b\u3055\u308c\u308b\u3002DrugB \u306e\u30aa\u30c3\u30ba\u6bd4\u306f\u307b\u307c1\u3067\u3042\u308a\u3001Drug A \u3068 B \u306f\u7570\u306a\u308b\u3068\u306f\u8a00\u3048\u306a\u304b\u3063\u305f\u3068\u3044\u3046\u7d50\u679c\u306b\u306a\u3063\u3066\u3044\u308b\u3002<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">\u30ab\u30a6\u30f3\u30c8\u30c7\u30fc\u30bf\u306e\u8a08\u7b97\u4f8b<\/h3>\n\n\n\n<p>\u30ab\u30a6\u30f3\u30c8\u30c7\u30fc\u30bf\uff08\u30a4\u30d9\u30f3\u30c8\u767a\u751f\u56de\u6570\u306a\u3069\u3001\u975e\u8ca0\u306e\u6574\u6570\u5024\uff09\u3092\u4e00\u822c\u5316\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\u3067\u89e3\u6790\u3059\u308b\u5834\u5408\u3001\u5fdc\u7b54\u5909\u6570\u304c\u30dd\u30a2\u30bd\u30f3\u5206\u5e03\u306b\u5f93\u3046\u3068\u4eee\u5b9a\u3059\u308b<strong>\u30dd\u30a2\u30bd\u30f3\u6df7\u5408\u30e2\u30c7\u30eb<\/strong>\u304c\u4e00\u822c\u7684\u3067\u3042\u308b\u3002\u305f\u3060\u3057\u3001\u30ab\u30a6\u30f3\u30c8\u30c7\u30fc\u30bf\u306b\u306f\u904e\u5206\u6563\uff08overdispersion\uff09\u3068\u547c\u3070\u308c\u308b\u73fe\u8c61\u304c\u3057\u3070\u3057\u3070\u898b\u3089\u308c\u308b\u3002\u3053\u308c\u306f\u3001\u30c7\u30fc\u30bf\u306e\u5206\u6563\u304c\u5e73\u5747\u3088\u308a\u306f\u308b\u304b\u306b\u5927\u304d\u3044\u5834\u5408\u306b\u8d77\u3053\u308a\u3001\u30dd\u30a2\u30bd\u30f3\u5206\u5e03\u306e\u4eee\u5b9a\uff08\u5e73\u5747\u3068\u5206\u6563\u304c\u7b49\u3057\u3044\uff09\u304c\u7834\u7dbb\u3057\u3066\u3044\u308b\u3053\u3068\u3092\u793a\u5506\u3057\u3066\u3044\u308b\u3002\u3053\u306e\u3088\u3046\u306a\u5834\u5408\u3001<strong>\u8ca0\u306e\u4e8c\u9805\u6df7\u5408\u30e2\u30c7\u30eb<\/strong>\u3092\u4f7f\u7528\u3059\u308b\u3053\u3068\u3067\u3001\u904e\u5206\u6563\u3092\u9069\u5207\u306b\u30e2\u30c7\u30eb\u5316\u3067\u304d\u308b\u3002<\/p>\n\n\n\n<p>R\u3067\u306f\u3001<code>lme4<\/code>\u30d1\u30c3\u30b1\u30fc\u30b8\u306e<code>glmer()<\/code>\u95a2\u6570\u304c\u30dd\u30a2\u30bd\u30f3\u6df7\u5408\u30e2\u30c7\u30eb\u3092\u3001<code>glmmTMB<\/code>\u30d1\u30c3\u30b1\u30fc\u30b8\u306e<code>glmmTMB()<\/code>\u95a2\u6570\u304c\u30dd\u30a2\u30bd\u30f3\u6df7\u5408\u30e2\u30c7\u30eb\u3068\u8ca0\u306e\u4e8c\u9805\u6df7\u5408\u30e2\u30c7\u30eb\u306e\u4e21\u65b9\u3092\u30b5\u30dd\u30fc\u30c8\u3057\u3066\u304a\u308a\u3001\u3088\u308a\u67d4\u8edf\u306a\u30e2\u30c7\u30eb\u69cb\u7bc9\u304c\u53ef\u80fd\u3067\u3042\u308b\u3002\u3053\u3053\u3067\u306f\u3001<code>glmmTMB<\/code>\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u4f7f\u7528\u3057\u305f\u4f8b\u3092\u793a\u3059\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code># \u5fc5\u8981\u306b\u5fdc\u3058\u3066\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\n# install.packages(\"glmmTMB\")\n# install.packages(\"dplyr\") # \u30c7\u30fc\u30bf\u64cd\u4f5c\u7528\n# install.packages(\"ggplot2\") # \u53ef\u8996\u5316\u7528\n\nlibrary(glmmTMB)\nlibrary(dplyr)\nlibrary(ggplot2)\n\n# \u4f8b\u3068\u3057\u3066\u3001\u67b6\u7a7a\u306e\u30c7\u30fc\u30bf\u3092\u751f\u6210\n# 50\u4eba\u306e\u5598\u606f\u60a3\u8005\u304c2\u3064\u306e\u6cbb\u7642\u7fa4\u306b\u5272\u308a\u4ed8\u3051\u3089\u308c\u3001\n# \u30d9\u30fc\u30b9\u30e9\u30a4\u30f3\u30011\u5e74\u76ee\u30012\u5e74\u76ee\u306e3\u6642\u70b9\u3067\u5e74\u9593\u5165\u9662\u56de\u6570\u3092\u8a18\u9332\nset.seed(1234) # \u518d\u73fe\u6027\u306e\u305f\u3081\u306e\u30b7\u30fc\u30c9\u8a2d\u5b9a\nn_patients &lt;- 50\nn_time_points &lt;- 3 # \u30d9\u30fc\u30b9\u30e9\u30a4\u30f3(0), 1\u5e74\u76ee(1), 2\u5e74\u76ee(2)\n\npatient_id_vec &lt;- rep(1:n_patients, each = n_time_points)\ntime_vec &lt;- rep(0:(n_time_points - 1), times = n_patients) # \u6642\u70b9: 0, 1, 2\n\n# \u60a3\u8005\u306b\u5272\u308a\u5f53\u3066\u3089\u308c\u308b\u56fa\u5b9a\u52b9\u679c\u306f\u6642\u9593\u3092\u901a\u3058\u3066\u4e0d\u5909\ntreatment_vec &lt;- sample(c(\"NewTreatment\", \"StandardCare\"), n_patients, replace = TRUE) %>% \n  rep(each = n_time_points)\nage_vec &lt;- round(rnorm(n_patients, mean = 45, sd = 10)) %>% \n  rep(each = n_time_points)\nsmoking_history_vec &lt;- sample(c(\"Non-Smoker\", \"Ex-Smoker\", \"Current-Smoker\"), n_patients, replace = TRUE, prob = c(0.5, 0.3, 0.2)) %>% \n  rep(each = n_time_points)\n\n# \u60a3\u8005\u3054\u3068\u306e\u30e9\u30f3\u30c0\u30e0\u306a\u52b9\u679c\uff08\u5207\u7247\uff09\u3092\u751f\u6210\nrandom_effect_patient &lt;- rnorm(n_patients, mean = 0, sd = 0.7) %>% # \u6a19\u6e96\u504f\u5dee\u3092\u5c11\u3057\u5927\u304d\u304f\u3057\u3066\u60a3\u8005\u5dee\u3092\u5f37\u8abf\n  rep(each = n_time_points)\n\n# \u5165\u9662\u56de\u6570\u306e\u771f\u306e\u5e73\u5747\uff08\u30dd\u30a2\u30bd\u30f3\u5206\u5e03\u306e\u03bb\uff09\u3092\u8a2d\u5b9a\n# \u5bfe\u6570\u30ea\u30f3\u30af\u95a2\u6570\u3092\u4f7f\u7528\u3059\u308b\u305f\u3081\u3001\u7dda\u5f62\u4e88\u6e2c\u5b50\u306b\u30e9\u30f3\u30c0\u30e0\u52b9\u679c\u3092\u52a0\u3048\u308b\n# \u65b0\u6cbb\u7642\u306e\u52b9\u679c\u306f1\u5e74\u76ee\u30012\u5e74\u76ee\u306b\u3088\u308a\u9855\u8457\u306b\u51fa\u308b\u3088\u3046\u306b\u3059\u308b\n# \u6642\u9593\u7d4c\u904e\u306b\u3088\u308b\u5165\u9662\u56de\u6570\u306e\u5909\u5316 (\u4f8b: \u5f90\u3005\u306b\u6e1b\u5c11)\nlog_lambda &lt;- \n  -1.0 +                                      # \u30d9\u30fc\u30b9\u30e9\u30a4\u30f3\u306e\u5bfe\u6570\u5e73\u5747\n  # \u56fa\u5b9a\u52b9\u679c\u306e\u5f71\u97ff\n  ifelse(treatment_vec == \"NewTreatment\" &amp; time_vec > 0, -0.6, 0) + # \u65b0\u6cbb\u7642\u306e\u52b9\u679c (\u6cbb\u7642\u5f8c\u306e\u307f)\n  0.02 * (age_vec - mean(age_vec)) +          # \u5e74\u9f62\u306e\u52b9\u679c (\u4e2d\u5fc3\u5316)\n  ifelse(smoking_history_vec == \"Ex-Smoker\", 0.4, 0) + # \u55ab\u7159\u6b74\u306e\u52b9\u679c\n  ifelse(smoking_history_vec == \"Current-Smoker\", 0.8, 0) +\n  # \u6642\u9593\u306e\u52b9\u679c (\u4f8b: \u6642\u9593\u304c\u7d4c\u3064\u3068\u5c11\u3057\u305a\u3064\u5165\u9662\u304c\u6e1b\u308b)\n  ifelse(time_vec == 1, -0.2, 0) +            # 1\u5e74\u76ee\u306e\u52b9\u679c\n  ifelse(time_vec == 2, -0.4, 0) +            # 2\u5e74\u76ee\u306e\u52b9\u679c\n  # \u60a3\u8005\u3054\u3068\u306e\u30e9\u30f3\u30c0\u30e0\u52b9\u679c\n  random_effect_patient\n\n# \u5e73\u5747\u5165\u9662\u56de\u6570\nlambda &lt;- exp(log_lambda)\n\n# \u5165\u9662\u56de\u6570\uff08\u30dd\u30a2\u30bd\u30f3\u5206\u5e03\u306b\u5f93\u3046\u3068\u4eee\u5b9a\uff09\u3092\u751f\u6210\n# \u904e\u5206\u6563\u3092\u30b7\u30df\u30e5\u30ec\u30fc\u30c8\u3059\u308b\u305f\u3081\u3001\u8ca0\u306e\u4e8c\u9805\u5206\u5e03\u3067\u751f\u6210\u3057\u3066\u307f\u308b (\u30dd\u30a2\u30bd\u30f3\u3067\u3084\u308a\u305f\u3051\u308c\u3070 rpois \u306b\u5909\u66f4)\nannual_hospitalizations &lt;- rnbinom(n_patients * n_time_points, size = 1.5, mu = lambda) # size\u30921.5\u306b\u8a2d\u5b9a\u3057\u3066\u5c11\u3057\u904e\u5206\u6563\u3092\u6301\u305f\u305b\u308b\n# annual_hospitalizations &lt;- rpois(n_patients * n_time_points, lambda)\n\n# \u30c7\u30fc\u30bf\u30d5\u30ec\u30fc\u30e0\u306e\u4f5c\u6210\ndf_repeated_count &lt;- data.frame(\n  patient_id = factor(patient_id_vec),\n  time = factor(time_vec, levels = 0:2, labels = c(\"Baseline\", \"Year1\", \"Year2\")), # \u6642\u70b9\u3092\u56e0\u5b50\u3068\u3057\u3066\u6307\u5b9a\n  treatment = factor(treatment_vec),\n  age = age_vec,\n  smoking_history = factor(smoking_history_vec, levels = c(\"Non-Smoker\", \"Ex-Smoker\", \"Current-Smoker\")),\n  annual_hospitalizations = annual_hospitalizations\n)\n\n# \u30c7\u30fc\u30bf\u306e\u78ba\u8a8d\nhead(df_repeated_count)\nstr(df_repeated_count)\nsummary(df_repeated_count$annual_hospitalizations)\ntable(df_repeated_count$time, df_repeated_count$treatment)\n\n# \u30dd\u30a2\u30bd\u30f3\u6df7\u5408\u30e2\u30c7\u30eb\u306e\u5b9f\u884c\n# time\u3092\u56fa\u5b9a\u52b9\u679c\u3068\u3057\u3066\u542b\u3081\u308b\n# (1 | patient_id): patient_id\u3054\u3068\u306e\u5207\u7247\u3092\u5909\u91cf\u52b9\u679c\u3068\u3057\u3066\u8003\u616e\nmodel_poisson_repeated &lt;- glmmTMB(annual_hospitalizations ~ treatment + age + smoking_history + time + (1 | patient_id),\n                                  data = df_repeated_count,\n                                  family = poisson(link = \"log\"))\n\n# \u30e2\u30c7\u30eb\u7d50\u679c\u306e\u8868\u793a\nsummary(model_poisson_repeated)\n\n# \u6b8b\u5dee\u306e\u78ba\u8a8d\uff08\u904e\u5206\u6563\u306e\u5146\u5019\u304c\u306a\u3044\u304b\uff09\n# \u30e2\u30c7\u30eb\u8a3a\u65ad\u30d7\u30ed\u30c3\u30c8\u306a\u3069\u3092\u884c\u3046\u3053\u3068\u304c\u63a8\u5968\u3055\u308c\u308b\n# plot(model_poisson_repeated) # glmmTMB\u306eplot\u30e1\u30bd\u30c3\u30c9\u306f\u3088\u308a\u8a73\u7d30\u306a\u8a3a\u65ad\u3092\u30b5\u30dd\u30fc\u30c8\n\n# \u3082\u3057\u904e\u5206\u6563\u304c\u7591\u308f\u308c\u308b\u5834\u5408\u306f\u3001\u8ca0\u306e\u4e8c\u9805\u6df7\u5408\u30e2\u30c7\u30eb\u3092\u4f7f\u7528\nmodel_negbin_repeated &lt;- glmmTMB(annual_hospitalizations ~ treatment + age + smoking_history + time + (1 | patient_id),\n                                 data = df_repeated_count,\n                                 family = nbinom2(link = \"log\"))\n\nsummary(model_negbin_repeated)\n\n# \u30e2\u30c7\u30eb\u306e\u6bd4\u8f03 (AIC\u5024)\nAIC(model_poisson_repeated, model_negbin_repeated)<\/code><\/pre>\n\n\n\n<p><strong>\u5b9f\u884c\u7d50\u679c\uff1a<\/strong><\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>> # \u30e2\u30c7\u30eb\u7d50\u679c\u306e\u8868\u793a\n> summary(model_poisson_repeated)\n Family: poisson  ( log )\nFormula:          \nannual_hospitalizations ~ treatment + age + smoking_history +  \n    time + (1 | patient_id)\nData: df_repeated_count\n\n      AIC       BIC    logLik -2*log(L)  df.resid \n    299.8     323.9    -141.9     283.8       142 \n\nRandom effects:\n\nConditional model:\n Groups     Name        Variance Std.Dev.\n patient_id (Intercept) 0.3296   0.5741  \nNumber of obs: 150, groups:  patient_id, 50\n\nConditional model:\n                              Estimate Std. Error z value Pr(>|z|)    \n(Intercept)                   -2.76770    0.70730  -3.913 9.11e-05 ***\ntreatmentStandardCare          0.56851    0.34141   1.665   0.0959 .  \nage                            0.02433    0.01330   1.829   0.0673 .  \nsmoking_historyEx-Smoker       0.58910    0.36131   1.630   0.1030    \nsmoking_historyCurrent-Smoker  0.85493    0.38553   2.218   0.0266 *  \ntimeYear1                     -0.17435    0.29601  -0.589   0.5558    \ntimeYear2                      0.07696    0.27756   0.277   0.7816    \n---\nSignif. codes:  0 \u2018***\u2019 0.001 \u2018**\u2019 0.01 \u2018*\u2019 0.05 \u2018.\u2019 0.1 \u2018 \u2019 1<\/code><\/pre>\n\n\n\n<p>treatmentStadardCare \u306e 0.56851 \u306f\u3001\u65b0\u6cbb\u7642\u7fa4\u3068\u6bd4\u8f03\u3057\u3066\u3001\u6a19\u6e96\u6cbb\u7642\u7fa4\u306f\u3001\u5e74\u9593\u5165\u9662\u56de\u6570\u304c\u3001\u5bfe\u6570\u5e73\u5747\u3067 0.56851 \u56de\u3001\u771f\u6570\u306b\u3059\u308b\u3068 exp(0.56851) = 1.765734\u3001\u3064\u307e\u308a\u3001\u7d04 1.8 \u56de\u591a\u3044\u3068\u63a8\u5b9a\u3055\u308c\u305f\u3002\u305f\u3060\u3057\u3001\u7d71\u8a08\u5b66\u7684\u6709\u610f\u3067\u306f\u306a\u304b\u3063\u305f\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>> # \u3082\u3057\u904e\u5206\u6563\u304c\u7591\u308f\u308c\u308b\u5834\u5408\u306f\u3001\u8ca0\u306e\u4e8c\u9805\u6df7\u5408\u30e2\u30c7\u30eb\u3092\u4f7f\u7528\n> model_negbin_repeated &lt;- glmmTMB(annual_hospitalizations ~ treatment + age + smoking_history + time + (1 | patient_id),\n+                                  data = df_repeated_count,\n+                                  family = nbinom2(link = \"log\"))\n> summary(model_negbin_repeated)\n Family: nbinom2  ( log )\nFormula:          \nannual_hospitalizations ~ treatment + age + smoking_history +  \n    time + (1 | patient_id)\nData: df_repeated_count\n\n      AIC       BIC    logLik -2*log(L)  df.resid \n    286.6     313.7    -134.3     268.6       141 \n\nRandom effects:\n\nConditional model:\n Groups     Name        Variance  Std.Dev. \n patient_id (Intercept) 1.754e-08 0.0001324\nNumber of obs: 150, groups:  patient_id, 50\n\nDispersion parameter for nbinom2 family (): 0.728 \n\nConditional model:\n                              Estimate Std. Error z value Pr(>|z|)    \n(Intercept)                   -2.51040    0.72986  -3.440 0.000583 ***\ntreatmentStandardCare          0.56779    0.35536   1.598 0.110085    \nage                            0.02244    0.01398   1.604 0.108622    \nsmoking_historyEx-Smoker       0.61934    0.37778   1.639 0.101125    \nsmoking_historyCurrent-Smoker  0.84247    0.39962   2.108 0.035014 *  \ntimeYear1                     -0.23412    0.39391  -0.594 0.552284    \ntimeYear2                      0.08020    0.37520   0.214 0.830750    \n---\nSignif. codes:  0 \u2018***\u2019 0.001 \u2018**\u2019 0.01 \u2018*\u2019 0.05 \u2018.\u2019 0.1 \u2018 \u2019 1<\/code><\/pre>\n\n\n\n<p>\u8ca0\u306e\u4e8c\u9805\u5206\u5e03\u3092\u60f3\u5b9a\u3057\u3066\u8a08\u7b97\u3059\u308b\u3068\u3001\u5bfe\u6570\u5e73\u5747\u3067\u30010.56779 \u3067\u3001\u771f\u6570\u306b\u3059\u308b\u3068\u3001exp(0.56779) = 1.764363 \u3068\u3053\u3061\u3089\u3082\u3001\u7d04 1.8 \u56de\u3001\u6a19\u6e96\u6cbb\u7642\u7fa4\u306e\u307b\u3046\u304c\u591a\u3044\u3068\u3044\u3046\u8a08\u7b97\u7d50\u679c\u3067\u3042\u3063\u305f\u3002\u3057\u304b\u3057\u3001\u3053\u3061\u3089\u3067\u3082\u7d71\u8a08\u5b66\u7684\u6709\u610f\u3067\u306f\u306a\u304b\u3063\u305f\u3002<\/p>\n\n\n\n<pre class=\"wp-block-code\"><code>\n> # \u30e2\u30c7\u30eb\u306e\u6bd4\u8f03 (AIC\u5024)\n> AIC(model_poisson_repeated, model_negbin_repeated)\n                       df      AIC\nmodel_poisson_repeated  8 299.8395\nmodel_negbin_repeated   9 286.5831<\/code><\/pre>\n\n\n\n<p>\u30dd\u30a2\u30bd\u30f3\u5206\u5e03\u3092\u4eee\u5b9a\u3057\u305f\u30e2\u30c7\u30eb\u3068\u8ca0\u306e\u4e8c\u9805\u5206\u5e03\u3092\u4eee\u5b9a\u3057\u305f\u30e2\u30c7\u30eb\u3092\u5f53\u3066\u306f\u307e\u308a\u306e\u89b3\u70b9\u304b\u3089 AIC \u3067\u6bd4\u8f03\u3059\u308b\u3068\u3001\u8ca0\u306e\u4e8c\u9805\u5206\u5e03\u3092\u4eee\u5b9a\u3057\u305f\u30e2\u30c7\u30eb\u306e\u307b\u3046\u304c\u3001\u82e5\u5e72 AIC \u304c\u4f4e\u304f\u3001\u6bd4\u8f03\u7684\u5f53\u3066\u306f\u307e\u308a\u306f\u3088\u3044\u3068\u306e\u7d50\u679c\u306b\u306a\u3063\u305f\u3002\u305f\u3060\u3057\u3001\u4e0a\u8ff0\u306e\u901a\u308a\u3001\u3069\u3061\u3089\u306e\u30e2\u30c7\u30eb\u3067\u3082\u3001\u65b0\u85ac\u6cbb\u7642\u7fa4\u3068\u6a19\u6e96\u6cbb\u7642\u7fa4\u306f\u3001\u7d71\u8a08\u5b66\u7684\u6709\u610f\u306a\u5dee\u306f\u306a\u304f\u3001\u4e21\u7fa4\u306b\u5dee\u304c\u3042\u308b\u3068\u306f\u8a00\u3048\u306a\u304b\u3063\u305f\u3002<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\u307e\u3068\u3081<\/h2>\n\n\n\n<p>\u4e00\u822c\u5316\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\uff08GLMM\uff09\u306f\u3001\u7dda\u5f62\u6df7\u5408\u30e2\u30c7\u30eb\u306e\u67a0\u7d44\u307f\u3092\u5909\u91cf\u52b9\u679c\u306b\u62e1\u5f35\u3057\u3064\u3064\u3001\u4e00\u822c\u5316\u7dda\u5f62\u30e2\u30c7\u30eb\u306e\u67d4\u8edf\u6027\u3092\u5229\u7528\u3057\u3066\u3001\u69d8\u3005\u306a\u5206\u5e03\u3092\u6301\u3064\u5fdc\u7b54\u5909\u6570\u306b\u5bfe\u5fdc\u3067\u304d\u308b\u5f37\u529b\u306a\u7d71\u8a08\u30e2\u30c7\u30ea\u30f3\u30b0\u30c4\u30fc\u30eb\u3067\u3042\u308b\u3002\u4e8c\u9805\u30a4\u30d9\u30f3\u30c8\u3001\u9806\u5e8f\u30ab\u30c6\u30b4\u30ea\u3001\u30ab\u30a6\u30f3\u30c8\u30c7\u30fc\u30bf\u306a\u3069\u3001\u591a\u69d8\u306a\u81e8\u5e8a\u7814\u7a76\u30c7\u30fc\u30bf\u306b\u304a\u3044\u3066\u3001\u500b\u4f53\u9593\u306e\u3070\u3089\u3064\u304d\u3084\u53cd\u5fa9\u6e2c\u5b9a\u306e\u76f8\u95a2\u69cb\u9020\u3092\u9069\u5207\u306b\u8003\u616e\u3057\u305f\u89e3\u6790\u3092\u53ef\u80fd\u306b\u3059\u308b\u3002GLMM\u3092\u7406\u89e3\u3057\u3001\u9069\u5207\u306b\u6d3b\u7528\u3059\u308b\u3053\u3068\u3067\u3001\u3088\u308a\u6b63\u78ba\u3067\u30ed\u30d0\u30b9\u30c8\u306a\u7814\u7a76\u7d50\u679c\u3092\u5f97\u308b\u3053\u3068\u304c\u3067\u304d\u3001\u81e8\u5e8a\u7684\u610f\u601d\u6c7a\u5b9a\u306b\u8ca2\u732e\u3059\u308b\u3053\u3068\u304c\u671f\u5f85\u3055\u308c\u308b\u3002R\u306a\u3069\u306e\u7d71\u8a08\u30bd\u30d5\u30c8\u30a6\u30a7\u30a2\u3092\u7528\u3044\u308b\u3053\u3068\u3067\u3001GLMM\u306f\u5bb9\u6613\u306b\u5b9f\u884c\u3067\u304d\u308b\u306e\u3067\u3001\u7d50\u679c\u306e\u89e3\u91c8\u304c\u9069\u5207\u306b\u3067\u304d\u308b\u3088\u3046\u306b\u3057\u3066\u3044\u304d\u305f\u3044\u3002<\/p>\n\n\n\n\n","protected":false},"excerpt":{"rendered":"<p>\u81e8\u5e8a\u7814\u7a76\u3084\u751f\u7269\u7d71\u8a08\u5b66\u306e\u5206\u91ce\u3067\u306f\u3001\u60a3\u8005\u3054\u3068\u306e\u3070\u3089\u3064\u304d\u3084\u6e2c\u5b9a\u306e\u53cd\u5fa9\u6027\u306a\u3069\u3001\u30c7\u30fc\u30bf\u304c\u6301\u3064\u8907\u96d1\u306a\u69cb\u9020\u3092\u8003\u616e\u3059\u308b\u3053\u3068\u304c\u4e0d\u53ef\u6b20\u3067\u3042\u308b\u3002\u3057\u304b\u3057\u3001\u57fa\u790e\u7684\u306a\u7d71\u8a08\u30e2\u30c7\u30eb\u3067\u306f\u3001\u3053\u306e\u3088\u3046\u306a\u8907\u96d1\u6027\u3092\u5341\u5206\u306b\u6349\u3048\u304d\u308c\u306a\u3044\u3002\u305d\u3053\u3067\u5fc5\u8981\u3068\u306a\u3063\u3066\u304f\u308b\u306e\u304c\u3001 [&hellip;]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"swell_btn_cv_data":"","_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[77],"tags":[],"class_list":["post-4014","post","type-post","status-publish","format-standard","hentry","category-77"],"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/posts\/4014","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/comments?post=4014"}],"version-history":[{"count":9,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/posts\/4014\/revisions"}],"predecessor-version":[{"id":4072,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/posts\/4014\/revisions\/4072"}],"wp:attachment":[{"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/media?parent=4014"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/categories?post=4014"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/tags?post=4014"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}