{"id":525,"date":"2018-08-04T17:41:02","date_gmt":"2018-08-04T08:41:02","guid":{"rendered":"https:\/\/best-biostatistics.com\/toukei-er\/entry\/how-to-determine-sample-size-in-logistic-regression-with-continuous-variable\/"},"modified":"2024-10-13T21:48:57","modified_gmt":"2024-10-13T12:48:57","slug":"how-to-determine-sample-size-in-logistic-regression-with-continuous-variable","status":"publish","type":"post","link":"https:\/\/best-biostatistics.com\/toukei-er\/entry\/how-to-determine-sample-size-in-logistic-regression-with-continuous-variable\/","title":{"rendered":"R \u3067\u8aac\u660e\u5909\u6570\u304c\u9023\u7d9a\u30c7\u30fc\u30bf\u306e\u5834\u5408\u306b\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u5206\u6790\u3067\u5fc5\u8981\u306a\u30b5\u30f3\u30d7\u30eb\u30b5\u30a4\u30ba\u3092\u8a08\u7b97\u3059\u308b\u65b9\u6cd5"},"content":{"rendered":"\n

\u72ec\u7acb\u5909\u6570\u304c\u9023\u7d9a\u91cf\u306e\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u306e\u30b5\u30f3\u30d7\u30eb\u30b5\u30a4\u30ba\u8a08\u7b97\u306e\u65b9\u6cd5<\/p>\n\n\n\n\n\n\n\n

\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u306e\u30b5\u30f3\u30d7\u30eb\u30b5\u30a4\u30ba\u8a08\u7b97\u30b9\u30af\u30ea\u30d7\u30c8 \u9023\u7d9a\u30c7\u30fc\u30bf\u306e\u5834\u5408<\/h2>\n\n\n\n

p0\u3092\u5e73\u5747\u5024\u30ec\u30d9\u30eb\u3067\u306e\u767a\u751f\u78ba\u7387\u3068\u3059\u308b\u3002<\/p>\n\n\n\n

p1\u3092\u5e73\u5747\u5024\uff0b1SD\uff08\u6a19\u6e96\u504f\u5dee\uff09\u30ec\u30d9\u30eb\u3067\u306e\u767a\u751f\u78ba\u7387\u3068\u3059\u308b\u3002<\/p>\n\n\n\n

\u30b5\u30f3\u30d7\u30eb\u30b5\u30a4\u30ba\u8a08\u7b97\u306f\u4ee5\u4e0b\u306e\u3088\u3046\u306a\u30b9\u30af\u30ea\u30d7\u30c8\u306b\u306a\u308b\u3002<\/p>\n\n\n\n

\u8a08\u7b97\u7d50\u679c\u306en\u306f\u5fc5\u8981\u306a\u5bfe\u8c61\u8005\u5168\u54e1\u306e\u6570\u3060\u3002<\/p>\n\n\n\n

sample.size.logistic <-<\/span> function<\/span>(<\/span>p0,<\/span> p1,<\/span>\nsig.level=<\/span>.05<\/span>,<\/span> power=<\/span>.8<\/span>,<\/span>alternative=<\/span>c<\/span>(<\/span>\"two.sided\"<\/span>,<\/span>\"one.sided\"<\/span>)){<\/span>\nalternative <-<\/span> match.arg<\/span>(<\/span>alternative)<\/span>\ntside <-<\/span> switch<\/span>(<\/span>alternative,<\/span> one.sided=<\/span>1<\/span>,<\/span> two.sided=<\/span>2<\/span>)<\/span>\nodds0 <-<\/span> p0\/<\/span>(<\/span>1<\/span>-<\/span>p0)<\/span>\nodds1 <-<\/span> p1\/<\/span>(<\/span>1<\/span>-<\/span>p1)<\/span>\ntheta <-<\/span> odds1\/<\/span>odds0\nlambda <-<\/span> log<\/span>(<\/span>theta)<\/span>\nZa <-<\/span> qnorm<\/span>(<\/span>sig.level\/<\/span>tside,<\/span> lower.tail=<\/span>FALSE<\/span>)<\/span>\nZb <-<\/span> qnorm<\/span>(<\/span>power)<\/span>\ndelta <-<\/span> (<\/span>1<\/span>+<\/span>(<\/span>1<\/span>+<\/span>lambda^<\/span>2<\/span>)<\/span>*<\/span>(<\/span>exp<\/span>(<\/span>5<\/span>*<\/span>lambda^<\/span>2<\/span>\/<\/span>4<\/span>)))<\/span>\/<\/span>(<\/span>1<\/span>+<\/span>exp<\/span>(<\/span>-<\/span>lambda^<\/span>2<\/span>\/<\/span>4<\/span>))<\/span>\nn <-<\/span> ((<\/span>Za+<\/span>Zb*<\/span>exp<\/span>(<\/span>-<\/span>lambda^<\/span>2<\/span>\/<\/span>4<\/span>))<\/span>^<\/span>2<\/span>*<\/span>(<\/span>1<\/span>+<\/span>2<\/span>*<\/span>p0*<\/span>delta))<\/span>\/<\/span>(<\/span>p0*<\/span>lambda^<\/span>2<\/span>)<\/span>\nNOTE<\/span> <-<\/span> \"n is size of entire cohort\"<\/span>\nMETHOD <-<\/span> \"Sample size, logistic reg. with cont. var.\"<\/span>\nstructure<\/span>(<\/span>list<\/span>(<\/span>n =<\/span> n,<\/span>\n\"Prob. in average level\"<\/span> =<\/span> p0,<\/span>\n\"Prob. in ave.+SD level\"<\/span> =<\/span> p1,<\/span>\nsig.level =<\/span> sig.level,<\/span>\npower =<\/span> power,<\/span>\nalternative =<\/span> alternative,<\/span> note =<\/span> NOTE<\/span>,<\/span>\nmethod =<\/span> METHOD),<\/span> class =<\/span> \"power.htest\"<\/span>)<\/span>\n}<\/span>\n<\/code><\/pre>\n\n\n\n
\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u306e\u30b5\u30f3\u30d7\u30eb\u30b5\u30a4\u30ba\u8a08\u7b97 \u9023\u7d9a\u30c7\u30fc\u30bf\u306e\u5834\u5408\u306e\u4f8b<\/h2>\n\n\n\n
\u4f8b\u3048\u3070\u3001\u5e73\u5747\u5024\u30ec\u30d9\u30eb\u30670.08\uff0b1SD\u30671.5\u500d\u306b\u306a\u308b\u3068\u4e88\u60f3\u3057\u3066\u8a08\u7b97\u3059\u308b\u3068\u3001\u30b3\u30db\u30fc\u30c8\u5168\u4f53\u3067569\u4f8b\u5fc5\u8981\u3068\u8a08\u7b97\u3055\u308c\u308b\u3002<\/p>\n\n\n\n
><\/span> sample.size.logistic<\/span>(<\/span>p0=<\/span>0.08<\/span>,<\/span> p1=<\/span>0.08<\/span>*<\/span>1.5<\/span>)<\/span>\nSample size,<\/span> logistic reg. with cont. var.\nn =<\/span> 568.7537<\/span>\nProb. in<\/span> average level =<\/span> 0.08<\/span>\nProb. in<\/span> ave.+<\/span>SD level =<\/span> 0.12<\/span>\nsig.level =<\/span> 0.05<\/span>\npower =<\/span> 0.8<\/span>\nalternative =<\/span> two.sided\nNOTE:<\/span> n is size of entire cohort\n<\/code><\/pre>\n\n\n\n
\u4f8b\u3048\u3070\u3001\u5e73\u5747\u5024\u30ec\u30d9\u30eb\u30670.2\u3067\u767a\u751f\u3057\u30011\u5358\u4f4d\u4e0a\u304c\u308b\u3054\u3068\u306b1.1\u500d\u30011SD\u304c5\u3060\u3068\u3057\u3066\u3001\uff0b1SD\u3067\u30011.15<\/sup> = 1.61 \u500d\u3001\u767a\u751f\u3059\u308b\u3068\u60f3\u5b9a\u3059\u308b\u3068\u3001154\u4f8b\u306e\u30b3\u30db\u30fc\u30c8\u3067\u3088\u3044\u3068\u3044\u3046\u8a08\u7b97\u306b\u306a\u308b\u3002<\/p>\n\n\n\n
><\/span> sample.size.logistic<\/span>(<\/span>p0=<\/span>0.2<\/span>,<\/span> p1=<\/span>0.2<\/span>*<\/span>1.61<\/span>)<\/span>\nSample size,<\/span> logistic reg. with cont. var.\nn =<\/span> 153.2653<\/span>\nProb. in<\/span> average level =<\/span> 0.2<\/span>\nProb. in<\/span> ave.+<\/span>SD level =<\/span> 0.322<\/span>\nsig.level =<\/span> 0.05<\/span>\npower =<\/span> 0.8<\/span>\nalternative =<\/span> two.sided\nNOTE:<\/span> n is size of entire cohort\n<\/code><\/pre>\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>
\u307e\u3068\u3081<\/h2>\n\n\n\n
\u30ed\u30b8\u30b9\u30c6\u30a3\u30c3\u30af\u56de\u5e30\u306e\u30b5\u30f3\u30d7\u30eb\u30b5\u30a4\u30ba\u8a08\u7b97\u3067\u3001\u8aac\u660e\u5909\u6570\u304c\u9023\u7d9a\u30c7\u30fc\u30bf\u306e\u5834\u5408\u3092\u89e3\u8aac\u3057\u305f\u3002<\/p>\n\n\n\n
\u53c2\u8003\u306b\u306a\u308c\u3070\u3002<\/p>\n\n\n\n
\u53c2\u8003\u66f8\u7c4d<\/h2>\n\n\n\n
\"\u30ab\u30c6\u30b4\u30ea\u30ab\u30eb\u30c7\u30fc\u30bf\u89e3\u6790\u5165\u9580\"<\/a><\/figure>\n\n\n\n
<\/div>\n\n\n\n
\n
\u30ab\u30c6\u30b4\u30ea\u30ab\u30eb\u30c7\u30fc\u30bf\u89e3\u6790\u5165\u9580<\/a><\/p>\n