{"id":7176,"date":"2025-10-27T10:41:19","date_gmt":"2025-10-27T01:41:19","guid":{"rendered":"https:\/\/blog.since2020.jp\/?p=7176"},"modified":"2025-10-27T10:41:19","modified_gmt":"2025-10-27T01:41:19","slug":"qualitative_regression","status":"publish","type":"post","link":"https:\/\/since2020.jp\/media\/qualitative_regression\/","title":{"rendered":"\u56de\u5e30\u5206\u6790\u3092\u6975\u3081\u308b(3\/4)\uff1a\u8cea\u7684\u56de\u5e30"},"content":{"rendered":"\n<p>\u3053\u308c\u307e\u3067\u9023\u7d9a\u5024\u3092\u76ee\u7684\u5909\u6570\u3068\u3059\u308b\u7dda\u5f62\u56de\u5e30\u306b\u3064\u3044\u3066\u5b66\u3093\u3067\u304d\u307e\u3057\u305f\u3002\u3057\u304b\u3057\u3001\u5b9f\u969b\u306e\u5206\u6790\u3067\u306f\u3001\u300c\u75c5\u6c17\u306e\u6709\u7121\u300d\u300c\u8cfc\u5165\u306e\u6709\u7121\u300d\u300c\u9078\u629e\u80a2\u306e\u7a2e\u985e\u300d\u306a\u3069\u3001\u30ab\u30c6\u30b4\u30ea\u30ab\u30eb\u306a\u76ee\u7684\u5909\u6570\u3092\u6271\u3046\u3053\u3068\u3082\u591a\u304f\u3042\u308a\u307e\u3059\u3002\u4eca\u56de\u306f\u3001\u305d\u3046\u3057\u305f\u8cea\u7684\u306a\u76ee\u7684\u5909\u6570\u306b\u5bfe\u3059\u308b\u56de\u5e30\u5206\u6790\u624b\u6cd5\u306b\u3064\u3044\u3066\u8a73\u3057\u304f\u89e3\u8aac\u3057\u307e\u3059\u3002<\/p>\n\n\n<h2>\u4e00\u822c\u5316\u7dda\u5f62\u30e2\u30c7\u30eb<\/h2>\n<p>\u4e00\u822c\u5316\u7dda\u5f62\u30e2\u30c7\u30eb\uff08Generalized Linear Model, GLM\uff09\u306f\u3001\u7dda\u5f62\u56de\u5e30\u3092\u62e1\u5f35\u3057\u3001\u69d8\u3005\u306a\u78ba\u7387\u5206\u5e03\u306b\u5f93\u3046\u76ee\u7684\u5909\u6570\u306b\u5bfe\u5fdc\u3067\u304d\u308b\u30d5\u30ec\u30fc\u30e0\u30ef\u30fc\u30af\u3067\u3059\u3002GLM\u306f\u4ee5\u4e0b\u306e3\u3064\u306e\u8981\u7d20\u304b\u3089\u69cb\u6210\u3055\u308c\u307e\u3059\uff1a<\/p>\r\n<ol>\r\n\t<li><strong>\u78ba\u7387\u5206\u5e03\uff08Random Component\uff09<\/strong><br \/>\r\n\u76ee\u7684\u5909\u6570\\((Y)\\)\u304c\u6307\u6570\u578b\u5206\u5e03\u65cf\u306b\u5c5e\u3059\u308b\u78ba\u7387\u5206\u5e03\u306b\u5f93\u3046\u3068\u4eee\u5b9a\u3057\u307e\u3059\u3002<\/li>\r\n\t<li><strong>\u7dda\u5f62\u4e88\u6e2c\u5b50\uff08Systematic Component\uff09<\/strong><br \/>\r\n\u8aac\u660e\u5909\u6570\u306e\u7dda\u5f62\u7d50\u5408\uff1a \\[\\eta = \\beta_0 + \\beta_1 x_1 + \\beta_2 x_2 + \\cdots + \\beta_p x_p\\]<\/li>\r\n\t<li><strong>\u30ea\u30f3\u30af\u95a2\u6570\uff08Link Function\uff09<\/strong><br \/>\r\n\u671f\u5f85\u5024\\((\\mu = E[Y])\\)\u3068\u7dda\u5f62\u4e88\u6e2c\u5b50\\((\\eta)\\)\u3092\u7d50\u3076\u95a2\u6570\uff1a \\[g(\\mu) = \\eta\\]<\/li>\r\n<\/ol>\n\n<h2>\u6307\u6570\u578b\u5206\u5e03\u65cf<\/h2>\n<p><strong>\u6307\u6570\u578b\u5206\u5e03\u65cf<\/strong>\u306f\u3001\u78ba\u7387\u5bc6\u5ea6\u95a2\u6570\uff08\u307e\u305f\u306f\u78ba\u7387\u8cea\u91cf\u95a2\u6570\uff09\u304c\u4ee5\u4e0b\u306e\u5f62\u3067\u8868\u3055\u308c\u308b\u5206\u5e03\u65cf\u3067\u3059\uff1a<\/p>\r\n<p>\\[f(y; \\theta, \\phi) = \\exp\\left(\\frac{y\\theta &#8211; b(\\theta)}{a(\\phi)} + c(y, \\phi)\\right)\\]<\/p>\r\n<p>\u3053\u3053\u3067\uff1a<\/p>\r\n<ul>\r\n\t<li>\\(\\theta\\)\uff1a\u81ea\u7136\u30d1\u30e9\u30e1\u30fc\u30bf<\/li>\r\n\t<li>\\(\\phi\\)\uff1a\u5c3a\u5ea6\u30d1\u30e9\u30e1\u30fc\u30bf<\/li>\r\n\t<li>\\(a(\\cdot)\\), \\(b(\\cdot)\\), \\(c(\\cdot)\\)\uff1a\u65e2\u77e5\u306e\u95a2\u6570<\/li>\r\n<\/ul>\r\n<p>\u4e3b\u306a\u6307\u6570\u578b\u5206\u5e03\u65cf\u306b\u306f\u4ee5\u4e0b\u304c\u3042\u308a\u307e\u3059\uff1a<\/p>\r\n<ul>\r\n\t<li><strong>\u6b63\u898f\u5206\u5e03<\/strong>\uff1a\u7dda\u5f62\u56de\u5e30<\/li>\r\n\t<li><strong>\u30d9\u30eb\u30cc\u30fc\u30a4\u5206\u5e03<\/strong>\uff1a\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30<\/li>\r\n\t<li><strong>\u30dd\u30a2\u30bd\u30f3\u5206\u5e03<\/strong>\uff1a\u30dd\u30a2\u30bd\u30f3\u56de\u5e30<\/li>\r\n\t<li><strong>\u30ac\u30f3\u30de\u5206\u5e03<\/strong>\uff1a\u30ac\u30f3\u30de\u56de\u5e30<\/li>\r\n<\/ul>\n\n<h2>\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u30e2\u30c7\u30eb<\/h2>\n<p><strong>\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u30e2\u30c7\u30eb<\/strong>\u306f\u3001\u4e8c\u5024\u306e\u76ee\u7684\u5909\u6570\uff080\u307e\u305f\u306f1\uff09\u3092\u6271\u3046\u6700\u3082\u4ee3\u8868\u7684\u306a\u8cea\u7684\u56de\u5e30\u624b\u6cd5\u3067\u3059\u3002<\/p>\r\n<b>\u30e2\u30c7\u30eb\u306e\u5b9a\u5f0f\u5316<\/b>\r\n<p>\u6210\u529f\u78ba\u7387\u3092\\(p = P(Y=1)\\)\u3068\u3059\u308b\u3068\u3001\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u30e2\u30c7\u30eb\u306f\uff1a<\/p>\r\n<p>\\[\\log\\left(\\frac{p}{1-p}\\right) = \\beta_0 + \\beta_1 x_1 + \\cdots + \\beta_p x_p\\]<\/p>\r\n<p>\u5de6\u8fba\u306e\\(\\log\\left(\\frac{p}{1-p}\\right)\\)\u306f<strong>\u30ed\u30b8\u30c3\u30c8<\/strong>\uff08logit\uff09\u3068\u547c\u3070\u308c\u3001\u3053\u308c\u304c\u30ea\u30f3\u30af\u95a2\u6570\u3068\u306a\u308a\u307e\u3059\u3002<\/p>\r\n<p>\u3053\u308c\u3092\\(p\\)\u306b\u3064\u3044\u3066\u89e3\u304f\u3068\uff1a<\/p>\r\n<p>\\[p = \\frac{\\exp(\\beta_0 + \\beta_1 x_1 + \\cdots + \\beta_p x_p)}{1 + \\exp(\\beta_0 + \\beta_1 x_1 + \\cdots + \\beta_p x_p)}\\]<\/p>\r\n<b>\u30aa\u30c3\u30ba\u6bd4<\/b>\r\n<p>\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u306e\u91cd\u8981\u306a\u6982\u5ff5\u3068\u3057\u3066<strong>\u30aa\u30c3\u30ba\u6bd4<\/strong>\u304c\u3042\u308a\u307e\u3059\u3002\u8aac\u660e\u5909\u6570\\(x_j\\)\u304c1\u5358\u4f4d\u5897\u52a0\u3057\u305f\u3068\u304d\u306e\u30aa\u30c3\u30ba\u6bd4\u306f\uff1a<\/p>\r\n<p>\\[OR_j = \\exp(\\beta_j)\\]<\/p>\r\n<p>\u3053\u308c\u306f\u3001\u4ed6\u306e\u5909\u6570\u3092\u56fa\u5b9a\u3057\u305f\u3068\u304d\u306e\u300c\u6210\u529f\u306e\u30aa\u30c3\u30ba\u300d\u306e\u6bd4\u3092\u8868\u3057\u307e\u3059\u3002<\/p>\r\n<b>\u6700\u5c24\u63a8\u5b9a<\/b>\r\n<p>\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u306e\u4fc2\u6570\u306f\u6700\u5c24\u63a8\u5b9a\u6cd5\u306b\u3088\u308a\u6c42\u3081\u3089\u308c\u307e\u3059\u3002\u5c24\u5ea6\u95a2\u6570\u306f\uff1a<\/p>\r\n<p>\\[L(\\boldsymbol{\\beta}) = \\prod_{i=1}^{n} p_i^{y_i}(1-p_i)^{1-y_i}\\]<\/p>\r\n<p>\u5bfe\u6570\u5c24\u5ea6\u3092\u6700\u5927\u5316\u3059\u308b\\(\\boldsymbol{\\beta}\\)\u3092\u53cd\u5fa9\u7684\u306b\u6c42\u3081\u307e\u3059\uff08\u901a\u5e38\u306f\u30cb\u30e5\u30fc\u30c8\u30f3\u30fb\u30e9\u30d5\u30bd\u30f3\u6cd5\uff09\u3002<\/p>\n\n<h2>\u30d7\u30ed\u30d3\u30c3\u30c8\u56de\u5e30\u30e2\u30c7\u30eb<\/h2>\n<p><strong>\u30d7\u30ed\u30d3\u30c3\u30c8\u56de\u5e30\u30e2\u30c7\u30eb<\/strong>\u306f\u3001\u6a19\u6e96\u6b63\u898f\u5206\u5e03\u306e\u7d2f\u7a4d\u5206\u5e03\u95a2\u6570\u3092\u30ea\u30f3\u30af\u95a2\u6570\u3068\u3059\u308b\u56de\u5e30\u30e2\u30c7\u30eb\u3067\u3059\uff1a<\/p>\r\n<p>\\[P(Y=1) = \\Phi(\\beta_0 + \\beta_1 x_1 + \\cdots + \\beta_p x_p)\\]<\/p>\r\n<p>\u3053\u3053\u3067\u3001\\(\\Phi(\\cdot)\\)\u306f\u6a19\u6e96\u6b63\u898f\u5206\u5e03\u306e\u7d2f\u7a4d\u5206\u5e03\u95a2\u6570\u3067\u3059\u3002<\/p>\r\n<p>\u30d7\u30ed\u30d3\u30c3\u30c8\u56de\u5e30\u3068\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u306e\u9055\u3044\uff1a<\/p>\r\n<ul>\r\n\t<li><strong>\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30<\/strong>\uff1a\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u5206\u5e03\uff08\u3088\u308a\u539a\u3044\u88fe\uff09<\/li>\r\n\t<li><strong>\u30d7\u30ed\u30d3\u30c3\u30c8\u56de\u5e30<\/strong>\uff1a\u6b63\u898f\u5206\u5e03\uff08\u3088\u308a\u8584\u3044\u88fe\uff09<\/li>\r\n<\/ul>\r\n<p>\u5b9f\u7528\u4e0a\u3001\u4e21\u8005\u306e\u7d50\u679c\u306f\u975e\u5e38\u306b\u4f3c\u3066\u3044\u307e\u3059\u3002<\/p>\n\n<h2>\u30dd\u30a2\u30bd\u30f3\u56de\u5e30\u30e2\u30c7\u30eb<\/h2>\n<p><strong>\u30dd\u30a2\u30bd\u30f3\u56de\u5e30\u30e2\u30c7\u30eb<\/strong>\u306f\u3001\u30ab\u30a6\u30f3\u30c8\u30c7\u30fc\u30bf\uff08\u975e\u8ca0\u6574\u6570\uff09\u3092\u76ee\u7684\u5909\u6570\u3068\u3059\u308b\u56de\u5e30\u30e2\u30c7\u30eb\u3067\u3059\u3002<\/p>\r\n<b>\u30e2\u30c7\u30eb\u306e\u5b9a\u5f0f\u5316<\/b>\r\n<p>\u76ee\u7684\u5909\u6570\\(Y\\)\u304c\u30dd\u30a2\u30bd\u30f3\u5206\u5e03\\(Poisson(\\lambda)\\)\u306b\u5f93\u3046\u3068\u304d\uff1a<\/p>\r\n<p>\\[\\log(\\lambda) = \\beta_0 + \\beta_1 x_1 + \\cdots + \\beta_p x_p\\]<\/p>\r\n<p>\u3053\u308c\u306b\u3088\u308a\u3001\u671f\u5f85\u5024\u306f\uff1a<\/p>\r\n<p>\\[E[Y] = \\lambda = \\exp(\\beta_0 + \\beta_1 x_1 + \\cdots + \\beta_p x_p)\\]<\/p>\r\n<b>\u30dd\u30a2\u30bd\u30f3\u56de\u5e30\u306e\u89e3\u91c8<\/b>\r\n<p>\u56de\u5e30\u4fc2\u6570\\(\\beta_j\\)\u306e\u89e3\u91c8\uff1a<\/p>\r\n<ul>\r\n\t<li>\\(\\exp(\\beta_j)\\)\uff1a\u8aac\u660e\u5909\u6570\\(x_j\\)\u304c1\u5358\u4f4d\u5897\u52a0\u3057\u305f\u3068\u304d\u306e\u671f\u5f85\u30ab\u30a6\u30f3\u30c8\u306e\u6bd4\uff08Rate Ratio\uff09<\/li>\r\n<\/ul>\r\n<b>\u904e\u5206\u6563\u3078\u306e\u5bfe\u51e6<\/b>\r\n<p>\u30dd\u30a2\u30bd\u30f3\u5206\u5e03\u3067\u306f\u5206\u6563=\u5e73\u5747\u3067\u3059\u304c\u3001\u5b9f\u30c7\u30fc\u30bf\u3067\u306f\u904e\u5206\u6563\uff08\u5206\u6563&gt;\u5e73\u5747\uff09\u304c\u751f\u3058\u308b\u3053\u3068\u304c\u3042\u308a\u307e\u3059\u3002\u3053\u306e\u5834\u5408\uff1a<\/p>\r\n<ul>\r\n\t<li><strong>\u6e96\u30dd\u30a2\u30bd\u30f3\u56de\u5e30<\/strong>\uff1a\u5206\u6563\u306e\u8abf\u6574<\/li>\r\n\t<li><strong>\u8ca0\u306e\u4e8c\u9805\u56de\u5e30<\/strong>\uff1a\u904e\u5206\u6563\u3092\u8a31\u5bb9\u3059\u308b\u5206\u5e03<\/li>\r\n<\/ul>\n\n<h2>\u305d\u306e\u4ed6\u306eGLM<\/h2>\n<b>\u591a\u9805\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30<\/b>\r\n<p>3\u3064\u4ee5\u4e0a\u306e\u30ab\u30c6\u30b4\u30ea\u3092\u6301\u3064\u76ee\u7684\u5909\u6570\u306b\u5bfe\u5fdc\uff1a<\/p>\r\n<p>\\[P(Y=k) = \\frac{\\exp(\\boldsymbol{x}^T\\boldsymbol{\\beta}_k)}{\\sum_{j=1}^{K}\\exp(\\boldsymbol{x}^T\\boldsymbol{\\beta}_j)}\\]<\/p>\r\n<b>\u9806\u5e8f\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30<\/b>\r\n<p>\u9806\u5e8f\u306e\u3042\u308b\u30ab\u30c6\u30b4\u30ea\u5909\u6570\uff08\u4f8b\uff1a\u6e80\u8db3\u5ea6\u8a55\u4fa1\uff09\u306b\u5bfe\u5fdc\uff1a<\/p>\r\n<p>\\[P(Y \\leq k) = \\frac{\\exp(\\alpha_k &#8211; \\boldsymbol{x}^T\\boldsymbol{\\beta})}{1 + \\exp(\\alpha_k &#8211; \\boldsymbol{x}^T\\boldsymbol{\\beta})}\\]<\/p>\n\n<h2>R\u306b\u3088\u308b\u5b9f\u88c5\uff1a\u9ad8\u8840\u5727\u4e88\u6e2c\u306e\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30<\/h2>\n<p>\u4ee5\u4e0b\u3067\u306f\u3001\u9ad8\u8840\u5727\u306e\u6709\u7121\u3092\u5e74\u9f62\u30fbBMI\u30fb\u55ab\u7159\u30fb\u904b\u52d5\u306a\u3069\u306e\u8981\u56e0\u3067\u4e88\u6e2c\u3059\u308b\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u30e2\u30c7\u30eb\u3092\u69cb\u7bc9\u3057\u307e\u3059\u3002<!-- notionvc: d57d8689-ce21-4b8d-8aae-cdfb4a640706 --><\/p>\r\n<div class=\"hcb_wrap\">\r\n<pre><code class=\"language-r\"># \u30b7\u30f3\u30d7\u30eb\u306a\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u5206\u6790\r\nset.seed(123)\r\nn &lt;- 500\r\n\r\n# \u30c7\u30fc\u30bf\u751f\u6210\r\ndata &lt;- data.frame(\r\n  age = rnorm(n, 50, 15),\r\n  bmi = rnorm(n, 25, 5),\r\n  smoking = factor(sample(c(\"No\", \"Yes\"), n, replace = TRUE)),\r\n  exercise = factor(sample(c(\"No\", \"Yes\"), n, replace = TRUE))\r\n)\r\n\r\n# \u9ad8\u8840\u5727\u306e\u751f\u6210\uff08\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u95a2\u6570\uff09\r\nlogit_p &lt;- -3 + 0.05 * data$age + 0.1 * data$bmi +\r\n           ifelse(data$smoking == \"Yes\", 0.8, 0) +\r\n           ifelse(data$exercise == \"Yes\", -0.6, 0)\r\n\r\ndata$hypertension &lt;- factor(rbinom(n, 1, plogis(logit_p)),\r\n                           labels = c(\"Normal\", \"Hypertension\"))\r\n\r\n# \u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u30e2\u30c7\u30eb\r\nmodel &lt;- glm(hypertension ~ age + bmi + smoking + exercise,\r\n             family = binomial, data = data)\r\n\r\n# \u7d50\u679c\u8868\u793a\r\nsummary(model)\r\n\r\n# \u30aa\u30c3\u30ba\u6bd4\u306e\u8a08\u7b97\r\nodds_ratios &lt;- exp(coef(model))\r\nprint(round(odds_ratios, 3))\r\n\r\n# \u4e88\u6e2c\u7cbe\u5ea6\r\npredicted &lt;- ifelse(fitted(model) &gt; 0.5, \"Hypertension\", \"Normal\")\r\naccuracy &lt;- mean(predicted == data$hypertension)\r\ncat(sprintf(\"\u4e88\u6e2c\u7cbe\u5ea6: %.3f\\\\n\", accuracy))\r\n\r\n# \u52b9\u679c\u306e\u53ef\u8996\u5316\r\npar(mfrow = c(1, 2))\r\n\r\n# BMI\u306e\u52b9\u679c\r\nbmi_seq &lt;- seq(15, 40, 1)\r\nbmi_pred &lt;- predict(model,\r\n                   newdata = data.frame(age = 50, bmi = bmi_seq,\r\n                                      smoking = \"No\", exercise = \"No\"),\r\n                   type = \"response\")\r\n\r\nplot(bmi_seq, bmi_pred, type = \"l\", col = \"blue\", lwd = 2,\r\n     main = \"BMI Effect\", xlab = \"BMI\", ylab = \"Probability\")\r\n\r\n# \u55ab\u7159\u30fb\u904b\u52d5\u306e\u52b9\u679c\r\ncombinations &lt;- expand.grid(smoking = c(\"No\", \"Yes\"), exercise = c(\"No\", \"Yes\"))\r\ncomb_pred &lt;- predict(model,\r\n                    newdata = data.frame(age = 50, bmi = 25, combinations),\r\n                    type = \"response\")\r\n\r\nbarplot(comb_pred,\r\n        names.arg = c(\"No Smoking\\\\nNo Exercise\", \"Smoking\\\\nNo Exercise\",\r\n                     \"No Smoking\\\\nExercise\", \"Smoking\\\\nExercise\"),\r\n        col = c(\"lightgreen\", \"orange\", \"lightblue\", \"red\"),\r\n        main = \"Lifestyle Effects\", ylab = \"Probability\")\r\n<\/code><\/pre>\r\n<\/div>\r\n<pre><strong>\u51fa\u529b\u7d50\u679c<\/strong><!-- notionvc: 941977ee-840e-496b-bee0-3ef3e9bc402b --><\/pre>\r\n<p><img decoding=\"async\" src=\"https:\/\/since2020.jp\/media\/wp-content\/uploads\/2025\/10\/ra3.png\" alt=\"\" width=\"411\" height=\"408\" class=\"alignnone size-full wp-image-7351 aligncenter\" srcset=\"https:\/\/since2020.jp\/media\/wp-content\/uploads\/2025\/10\/ra3.png 411w, https:\/\/since2020.jp\/media\/wp-content\/uploads\/2025\/10\/ra3-300x298.png 300w, https:\/\/since2020.jp\/media\/wp-content\/uploads\/2025\/10\/ra3-150x150.png 150w, https:\/\/since2020.jp\/media\/wp-content\/uploads\/2025\/10\/ra3-120x120.png 120w\" sizes=\"(max-width: 411px) 100vw, 411px\" \/><\/p>\r\n<p><!-- notionvc: 912fe7f5-2c1b-40e3-bc15-1c30eebf9978 --><\/p>\n\n<h2>\u51fa\u529b\u7d50\u679c\u306e\u89e3\u91c8<\/h2>\n<p>\u5206\u6790\u7d50\u679c\u304b\u3089\u4ee5\u4e0b\u306e\u77e5\u898b\u304c\u5f97\u3089\u308c\u307e\u3059\uff1a<\/p>\r\n<b><strong>\u30aa\u30c3\u30ba\u6bd4\u306e\u89e3\u91c8\uff1a<\/strong><\/b>\r\n<ul>\r\n\t<li><strong>\u5e74\u9f62<\/strong>: \u30aa\u30c3\u30ba\u6bd4 \u2248 1.03 \u2192 \u5e74\u9f62\u304c\u9ad8\u3044\u307b\u3069\u9ad8\u8840\u5727\u30ea\u30b9\u30af\u5897\u52a0<\/li>\r\n\t<li><strong>BMI<\/strong>: \u30aa\u30c3\u30ba\u6bd4 \u2248 1.10 \u2192 BMI\u304c\u9ad8\u3044\u307b\u3069\u9ad8\u8840\u5727\u30ea\u30b9\u30af\u5897\u52a0<\/li>\r\n\t<li><strong>\u55ab\u7159<\/strong>: \u30aa\u30c3\u30ba\u6bd4 \u2248 2.21 \u2192 \u55ab\u7159\u306b\u3088\u308a\u9ad8\u8840\u5727\u30ea\u30b9\u30af\u304c\u7d042.2\u500d\u306b\u5897\u52a0<\/li>\r\n\t<li><strong>\u904b\u52d5<\/strong>: \u30aa\u30c3\u30ba\u6bd4 \u2248 0.44 \u2192 \u904b\u52d5\u306b\u3088\u308a\u9ad8\u8840\u5727\u30ea\u30b9\u30af\u304c\u7d04\u534a\u5206\u306b\u6e1b\u5c11<\/li>\r\n<\/ul>\r\n<b><strong>\u53ef\u8996\u5316\u7d50\u679c\u306e\u89e3\u91c8\uff1a<\/strong><\/b>\r\n<ul>\r\n\t<li><strong>BMI\u52b9\u679c\u30b0\u30e9\u30d5<\/strong>\uff1aBMI\u306e\u5897\u52a0\u3068\u3068\u3082\u306b\u9ad8\u8840\u5727\u78ba\u7387\u304c\u975e\u7dda\u5f62\u306b\u4e0a\u6607<\/li>\r\n\t<li><strong>\u751f\u6d3b\u7fd2\u6163\u52b9\u679c\u30b0\u30e9\u30d5<\/strong>\uff1a\u300c\u55ab\u7159\u306a\u3057\u30fb\u904b\u52d5\u3042\u308a\u300d\u304c\u6700\u3082\u4f4e\u30ea\u30b9\u30af\u3001\u300c\u55ab\u7159\u3042\u308a\u30fb\u904b\u52d5\u306a\u3057\u300d\u304c\u6700\u3082\u9ad8\u30ea\u30b9\u30af<\/li>\r\n<\/ul>\r\n<b><strong>\u30e2\u30c7\u30eb\u306e\u59a5\u5f53\u6027\u8a55\u4fa1\uff1a<\/strong><\/b>\r\n<ul>\r\n\t<li><strong>\u4e88\u6e2c\u7cbe\u5ea6<\/strong>: 88.2%\u3068\u3044\u3046\u9ad8\u3044\u7cbe\u5ea6\u3092\u9054\u6210<\/li>\r\n\t<li><strong>\u7d71\u8a08\u7684\u6709\u610f\u6027<\/strong>: \u3059\u3079\u3066\u306e\u5909\u6570\u304c\u6709\u610f\uff08p &lt; 0.05\uff09<\/li>\r\n\t<li><strong>AIC<\/strong>: 340.06\uff08\u30e2\u30c7\u30eb\u6bd4\u8f03\u306e\u6307\u6a19\uff09<\/li>\r\n<\/ul>\n\n<h2>\u307e\u3068\u3081<\/h2>\n<p>\u8cea\u7684\u56de\u5e30\u5206\u6790\u3001\u7279\u306b\u4e00\u822c\u5316\u7dda\u5f62\u30e2\u30c7\u30eb\u306b\u3064\u3044\u3066\u5b66\u3073\u307e\u3057\u305f\u3002\u91cd\u8981\u306a\u30dd\u30a4\u30f3\u30c8\u3092\u4ee5\u4e0b\u306b\u8a18\u3057\u307e\u3059\uff1a<\/p>\r\n<ol>\r\n\t<li><strong>\u4e00\u822c\u5316\u7dda\u5f62\u30e2\u30c7\u30eb<\/strong>\u306f\u7dda\u5f62\u56de\u5e30\u3092\u69d8\u3005\u306a\u78ba\u7387\u5206\u5e03\u306b\u62e1\u5f35<\/li>\r\n\t<li><strong>\u6307\u6570\u578b\u5206\u5e03\u65cf<\/strong>\u304c\u7406\u8ad6\u7684\u57fa\u76e4<\/li>\r\n\t<li><strong>\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30<\/strong>\u306f\u4e8c\u5024\u5206\u985e\u306e\u6a19\u6e96\u624b\u6cd5<\/li>\r\n\t<li><strong>\u30d7\u30ed\u30d3\u30c3\u30c8\u56de\u5e30<\/strong>\u306f\u6b63\u898f\u5206\u5e03\u30d9\u30fc\u30b9\u306e\u4ee3\u66ff\u624b\u6cd5<\/li>\r\n\t<li><strong>\u30dd\u30a2\u30bd\u30f3\u56de\u5e30<\/strong>\u306f\u30ab\u30a6\u30f3\u30c8\u30c7\u30fc\u30bf\u306b\u9069\u7528<\/li>\r\n\t<li><strong>\u30aa\u30c3\u30ba\u6bd4<\/strong>\u306b\u3088\u308b\u52b9\u679c\u306e\u89e3\u91c8\u304c\u91cd\u8981<\/li>\r\n\t<li>\u30e2\u30c7\u30eb\u8a3a\u65ad\u3068\u4e88\u6e2c\u6027\u80fd\u8a55\u4fa1\u304c\u5fc5\u9808<\/li>\r\n<\/ol>\r\n<p>\u6b21\u56de\u306f\u56de\u5e30\u5206\u6790\u306e\u7279\u6b8a\u306a\u5fdc\u7528\u3068\u3057\u3066\u3001\u751f\u5b58\u6642\u9593\u5206\u6790\u3084\u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u30e2\u30c7\u30eb\u306b\u3064\u3044\u3066\u5b66\u7fd2\u3057\u307e\u3059\u3002<\/p>","protected":false},"excerpt":{"rendered":"<p>\u3053\u308c\u307e\u3067\u9023\u7d9a\u5024\u3092\u76ee\u7684\u5909\u6570\u3068\u3059\u308b\u7dda\u5f62\u56de\u5e30\u306b\u3064\u3044\u3066\u5b66\u3093\u3067\u304d\u307e\u3057\u305f\u3002\u3057\u304b\u3057\u3001\u5b9f\u969b\u306e\u5206\u6790\u3067\u306f\u3001\u300c\u75c5\u6c17\u306e\u6709\u7121\u300d\u300c\u8cfc\u5165\u306e\u6709\u7121\u300d\u300c\u9078\u629e\u80a2\u306e\u7a2e\u985e\u300d\u306a\u3069\u3001\u30ab\u30c6\u30b4\u30ea\u30ab\u30eb\u306a\u76ee\u7684\u5909\u6570\u3092\u6271\u3046\u3053\u3068\u3082\u591a\u304f\u3042\u308a\u307e\u3059\u3002\u4eca\u56de\u306f\u3001\u305d\u3046\u3057\u305f\u8cea\u7684\u306a\u76ee\u7684\u5909\u6570\u306b\u5bfe\u3059 [&hellip;]<\/p>\n","protected":false},"author":89,"featured_media":7137,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"content-type":"","swell_btn_cv_data":"","footnotes":"","_wp_rev_ctl_limit":""},"categories":[1246],"tags":[894,484,57],"class_list":["post-7176","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-data-infrastructure","tag-r","tag-484","tag-57"],"_links":{"self":[{"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/posts\/7176","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/users\/89"}],"replies":[{"embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/comments?post=7176"}],"version-history":[{"count":1,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/posts\/7176\/revisions"}],"predecessor-version":[{"id":7487,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/posts\/7176\/revisions\/7487"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/media\/7137"}],"wp:attachment":[{"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/media?parent=7176"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/categories?post=7176"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/since2020.jp\/media\/wp-json\/wp\/v2\/tags?post=7176"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}