{"id":561,"date":"2018-06-28T00:00:48","date_gmt":"2018-06-27T15:00:48","guid":{"rendered":"https:\/\/best-biostatistics.com\/toukei-er\/?p=561"},"modified":"2025-02-10T19:50:24","modified_gmt":"2025-02-10T10:50:24","slug":"pharmacoeconomics-by-r","status":"publish","type":"post","link":"https:\/\/best-biostatistics.com\/toukei-er\/entry\/pharmacoeconomics-by-r\/","title":{"rendered":"R \u3067\u85ac\u5264\u7d4c\u6e08\u5206\u6790\u3092\u884c\u3046\u65b9\u6cd5"},"content":{"rendered":"\n

\u85ac\u5264\u7d4c\u6e08\u5b66\u306f\u3001\u85ac\u5264\u306e\u6301\u3064\u52b9\u529b\u3092\u3001\u7d4c\u6e08\u3067\u6e2c\u3063\u3066\u541f\u5473\u3059\u308b\u5b66\u554f\u3002<\/p>\n\n\n\n

\u65b0\u85ac\u306b\u3064\u3051\u305f\u85ac\u4fa1\u304c\u3001\u672c\u5f53\u306b\u305d\u306e\u85ac\u4fa1\u306b\u898b\u5408\u3063\u305f\u52b9\u679c\u3092\u767a\u63ee\u3057\u3066\u3044\u308b\u304b\u3092\u541f\u5473\u3057\u3001\u653f\u7b56\u306b\u751f\u304b\u305d\u3046\u3068\u3059\u308b\u5b66\u554f\u3002<\/p>\n\n\n\n\n\n\n\n

\u85ac\u5264\u7d4c\u6e08\u5b66\u306e\u66f8\u7c4d\u306b\u305d\u3063\u3066\u5b9f\u8df5\u3057\u3066\u307f\u308b<\/h2>\n\n\n\n

\u30cd\u30bf\u5143\u306f\u3001\u5742\u5dfb\u5f18\u4e4b \u8457\u3000\u3084\u3055\u3057\u304f\u5b66\u3076\u85ac\u5264\u7d4c\u6e08\u5b66<\/a><\/p>\n\n\n\n

\u672c\u7a3f\u306f\u3001\u30cd\u30bf\u672c\u306e\u5185\u5bb9\u3092\u5199\u7d4c\u3057\u305f\u306b\u904e\u304e\u306a\u3044\u306e\u3067\u3001\u8a73\u7d30\u306b\u3064\u3044\u3066\u306f\u30cd\u30bf\u672c\u3092\u78ba\u8a8d\u3055\u308c\u305f\u3044\u3002<\/p>\n\n\n\n

\u5224\u65ad\u5206\u6790\u30e2\u30c7\u30eb Decision analysis model \u305d\u306e1\uff08Exercise 5\uff09<\/h2>\n\n\n\n

\u53c2\u8003\u6587\u732e\uff1a<\/p>\n\n\n\n

Evans KW, et al: Pharmacoeconomics 1997;12(5):567-577<\/a>.<\/p>\n\n\n\n

\u6761\u4ef6\u8a2d\u5b9a\u3057\u305f\u306e\u3061\u306b\u3001<\/p>\n\n\n\n

    \n
  1. “\u6709\u52b9-\u5b8c\u5168\u6d88\u5931”<\/li>\n\n\n\n
  2. “\u6709\u52b9-\u518d\u767a-\u6d88\u5931”<\/li>\n\n\n\n
  3. “\u7121\u52b9-\u653e\u7f6e-\u6d88\u5931”<\/li>\n\n\n\n
  4. “\u7121\u52b9-\u7dca\u6025\u5165\u9662-\u81ea\u7136\u6d88\u5931”<\/li>\n\n\n\n
  5. “\u7121\u52b9-\u7dca\u6025\u5165\u9662-\u52a0\u7642-\u6d88\u5931”<\/li>\n<\/ol>\n\n\n\n

    \u3068\u3044\u30465\u3064\u306e\u30b7\u30ca\u30ea\u30aa\u306b\u3064\u3044\u3066\u3001\u52b9\u7528\u5024\u3068\u30b3\u30b9\u30c8\u3092\u898b\u7a4d\u3082\u308a\u3001\u8cbb\u7528\u5bfe\u52b9\u679c\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n

    \u305d\u3057\u3066\u3001C\/F\u7fa4\u3068A\u7fa4\u3068\u306e\u6bd4\u8f03\u3092\u884c\u3046\u3002<\/p>\n\n\n\n

    # C\/F\u7fa4<\/span>\nCF.eff <-<\/span> 0.38<\/span> #\u6709\u52b9\u7387<\/span>\nCF.rec <-<\/span> 0.30<\/span> #\u518d\u767a\u7387<\/span>\nCF.ineff <-<\/span> 1-0.38<\/span> #\u7121\u52b9\u7387<\/span>\nCF.ineff.ER <-<\/span> 0.08<\/span> #\u7121\u52b9\u6642\u7dca\u6025\u5165\u9662\u7387<\/span>\nCF.ineff.ER.eff <-<\/span> 0.998<\/span> #\u7121\u52b9\u6642\u7dca\u6025\u5165\u9662\u5f8c\u81ea\u7136\u6d88\u5931\u7387<\/span>\n# \u521d\u671f\u6cbb\u7642 \u6709\u52b9\u7387<\/span>\np1.CF <-<\/span> 0.38<\/span>\np1.AA <-<\/span> 0.56<\/span>\n# \u521d\u671f\u6cbb\u7642 \u7121\u52b9\u7387<\/span>\nq1.CF <-<\/span> 0.62<\/span>\nq1.AA <-<\/span> 0.44<\/span>\n# \u518d\u767a\u7387 \u518d\u767a\u306a\u304f\u5b8c\u5168\u6d88\u5931\u306e\u5272\u5408<\/span>\np2.CF <-<\/span> 0.70<\/span>\np2.AA <-<\/span> 0.60<\/span>\n# \u518d\u767a\u7387 \u518d\u767a \u540c\u6cbb\u7642\u5f8c\u6d88\u5931\u306e\u5272\u5408<\/span>\nq2.CF <-<\/span> 0.30<\/span>\nq2.AA <-<\/span> 0.40<\/span>\n# \u7121\u52b9\u6642 \u7dca\u6025\u5165\u9662\u7387<\/span>\np3.CF <-<\/span> 0.08<\/span>\np3.AA <-<\/span> 0.08<\/span>\n# \u7121\u52b9\u6642 \u653e\u7f6e \u81ea\u7136\u6d88\u5931\u306e\u5272\u5408<\/span>\nq3.CF <-<\/span> 0.92<\/span>\nq3.AA <-<\/span> 0.92<\/span>\n# \u5165\u9662\u5f8c\u7d4c\u904e \u81ea\u7136\u6d88\u5931\u306e\u5272\u5408<\/span>\np4.CF <-<\/span> 0.998<\/span>\np4.AA <-<\/span> 0.998<\/span>\n# \u5165\u9662\u5f8c\u7d4c\u904e \u5165\u9662\u52a0\u7642\u3067\u6d88\u5931\u3059\u308b\u5272\u5408<\/span>\nq4.CF <-<\/span> 0.002<\/span>\nq4.AA <-<\/span> 0.002<\/span>\n# \u521d\u671f\u6cbb\u7642\u306e\u8cbb\u7528<\/span>\nC1.CF <-<\/span> 1.32<\/span>\nC1.AA <-<\/span> 16.10<\/span>\n# \u518d\u767a\u6cbb\u7642\uff08\u540c\u6cbb\u7642\uff09\u306e\u8cbb\u7528<\/span>\nC2.CF <-<\/span> 1.32<\/span>\nC2.AA <-<\/span> 16.10<\/span>\n# \u7dca\u6025\u5165\u9662\u306e\u8cbb\u7528<\/span>\nC3.CF <-<\/span> 63.13<\/span>\nC3.AA <-<\/span> 63.13<\/span>\n# \u5165\u9662\u52a0\u7642\u306e\u8cbb\u7528<\/span>\nC4.CF <-<\/span> 1093.00<\/span>\nC4.AA <-<\/span> 1093.00<\/span>\nscenari <-<\/span> c(<\/span>\n\"\u6709\u52b9-\u5b8c\u5168\u6d88\u5931\",<\/span>\n\"\u6709\u52b9-\u518d\u767a-\u6d88\u5931\",<\/span>\n\"\u7121\u52b9-\u653e\u7f6e-\u6d88\u5931\",<\/span>\n\"\u7121\u52b9-\u7dca\u6025\u5165\u9662-\u81ea\u7136\u6d88\u5931\",<\/span>\n\"\u7121\u52b9-\u7dca\u6025\u5165\u9662-\u52a0\u7642-\u6d88\u5931\")<\/span>\nutility <-<\/span> c(1.0,<\/span> 0.9,<\/span> -0.3,<\/span> 0.1,<\/span> -0.3)<\/span> #\u52b9\u7528\u5024\u306e\u5272\u308a\u5f53\u3066<\/span><\/code><\/pre>\n\n\n\n
    \u307e\u305aC\/F\u7fa4\u306e\u8a08\u7b97\u3002<\/p>\n\n\n\n
    prob.CF <- c(\n  p1.CF*p2.CF,\n  p1.CF*q2.CF,\n  q1.CF*q3.CF,\n  q1.CF*p3.CF*p4.CF,\n  q1.CF*p3.CF*q4.CF)\ncost.CF <- c(\n  C1.CF,\n  C1.CF+C2.CF,\n  C1.CF,\n  C1.CF+C3.CF,\n  C1.CF+C4.CF)\ndata.frame(scenari, utility, prob.CF, cost.CF)<\/code><\/pre>\n\n\n\n
    \u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u8868\u793a\u3002<\/p>\n\n\n\n
    ><\/span> data.frame(<\/span>scenari,<\/span> utility,<\/span> prob.CF,<\/span> cost.CF)<\/span>\nscenari utility   prob.CF cost.CF\n1<\/span>           \u6709\u52b9-<\/span>\u5b8c\u5168\u6d88\u5931     1.0<\/span> 0.2660000<\/span>    1.32<\/span>\n2<\/span>          \u6709\u52b9-<\/span>\u518d\u767a-<\/span>\u6d88\u5931     0.9<\/span> 0.1140000<\/span>    2.64<\/span>\n3<\/span>          \u7121\u52b9-<\/span>\u653e\u7f6e-<\/span>\u6d88\u5931    -0.3<\/span> 0.5704000<\/span>    1.32<\/span>\n4<\/span>  \u7121\u52b9-<\/span>\u7dca\u6025\u5165\u9662-<\/span>\u81ea\u7136\u6d88\u5931     0.1<\/span> 0.0495008<\/span>   64.45<\/span>\n5<\/span> \u7121\u52b9-<\/span>\u7dca\u6025\u5165\u9662-<\/span>\u52a0\u7642-<\/span>\u6d88\u5931    -0.3<\/span> 0.0000992<\/span> 1094.32<\/span><\/code><\/pre>\n\n\n\n
    \u52b9\u7528\u306e\u671f\u5f85\u5024\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
    est.util.CF <-<\/span> prob.CF*<\/span>utility\nsum(<\/span>est.util.CF)<\/span><\/code><\/pre>\n\n\n\n
    \u52b9\u7528\u306e\u671f\u5f85\u5024\u306e\u5408\u8a08\u306f\u7d040.20\u3068\u8a08\u7b97\u3055\u308c\u308b\u3002<\/p>\n\n\n\n
    ><\/span> sum(<\/span>est.util.CF)<\/span>\n[1]<\/span> 0.2024003<\/span><\/code><\/pre>\n\n\n\n
    \u8cbb\u7528\u306e\u671f\u5f85\u5024\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
    est.cost.CF <-<\/span> prob.CF*<\/span>cost.CF\nsum(<\/span>est.cost.CF)<\/span><\/code><\/pre>\n\n\n\n
    \u8cbb\u7528\u306e\u671f\u5f85\u5024\u306e\u5408\u8a08\u306f\u7d044.7\u3068\u8a08\u7b97\u3055\u308c\u308b\u3002<\/p>\n\n\n\n
    ><\/span> sum(<\/span>est.cost.CF)<\/span>\n[1]<\/span> 4.703891<\/span><\/code><\/pre>\n\n\n\n
    \u8cbb\u7528\u306e\u671f\u5f85\u5024\u306e\u5408\u8a08\u3068\u52b9\u7528\u306e\u671f\u5f85\u5024\u306e\u5408\u8a08\u306e\u6bd4\u3092\u5e73\u5747\u8cbb\u7528\/\u52b9\u7528\u6bd4\u3068\u8a00\u3046\u3002<\/p>\n\n\n\n
    ><\/span> sum(<\/span>est.cost.CF)\/sum(<\/span>est.util.CF)<\/span>\n[1]<\/span> 23.24053<\/span><\/code><\/pre>\n\n\n\n
    \u6b21\u306bA\u7fa4\u306e\u8a08\u7b97\u3002<\/p>\n\n\n\n
    prob.AA <- c(\n  p1.AA*p2.AA,\n  p1.AA*q2.AA,\n  q1.AA*q3.AA,\n  q1.AA*p3.AA*p4.AA,\n  q1.AA*p3.AA*q4.AA)\ncost.AA <- c(\n  C1.AA,\n  C1.AA+C2.AA,\n  C1.AA,\n  C1.AA+C3.AA,\n  C1.AA+C4.AA)\ndata.frame(scenari, utility, prob.AA, cost.AA)<\/code><\/pre>\n\n\n\n
    \u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u8868\u793a\u3059\u308b\u3002<\/p>\n\n\n\n
    ><\/span> data.frame(<\/span>scenari,<\/span> utility,<\/span> prob.AA,<\/span> cost.AA)<\/span>\nscenari utility   prob.AA cost.AA\n1<\/span>           \u6709\u52b9-<\/span>\u5b8c\u5168\u6d88\u5931     1.0<\/span> 0.3360000<\/span>   16.10<\/span>\n2<\/span>          \u6709\u52b9-<\/span>\u518d\u767a-<\/span>\u6d88\u5931     0.9<\/span> 0.2240000<\/span>   32.20<\/span>\n3<\/span>          \u7121\u52b9-<\/span>\u653e\u7f6e-<\/span>\u6d88\u5931    -0.3<\/span> 0.4048000<\/span>   16.10<\/span>\n4<\/span>  \u7121\u52b9-<\/span>\u7dca\u6025\u5165\u9662-<\/span>\u81ea\u7136\u6d88\u5931     0.1<\/span> 0.0351296<\/span>   79.23<\/span>\n5<\/span> \u7121\u52b9-<\/span>\u7dca\u6025\u5165\u9662-<\/span>\u52a0\u7642-<\/span>\u6d88\u5931    -0.3<\/span> 0.0000704<\/span> 1109.10<\/span>\n<\/code><\/pre>\n\n\n\n
    \u52b9\u7528\u306e\u671f\u5f85\u5024\u5408\u8a08\u3001\u8cbb\u7528\u306e\u671f\u5f85\u5024\u5408\u8a08\u3092\u8a08\u7b97\u3057\u3001\u8cbb\u7528\u5bfe\u52b9\u7528\u306e\u6bd4\u3092\u53d6\u308b\u3002<\/p>\n\n\n\n
    # \u52b9\u7528\u306e\u671f\u5f85\u5024<\/span>\nest.util.AA <-<\/span> prob.AA*<\/span>utility\n# \u8cbb\u7528\u306e\u671f\u5f85\u5024<\/span>\nest.cost.AA <-<\/span> prob.AA*<\/span>cost.AA\n# \u5e73\u5747\u8cbb\u7528\/\u52b9\u7528\u6bd4<\/span>\nsum(<\/span>est.cost.AA)\/sum(<\/span>est.util.AA)<\/span>\n<\/code><\/pre>\n\n\n\n
    \u8cbb\u7528\u3068\u52b9\u7528\u306e\u671f\u5f85\u5024\u5408\u8a08\u540c\u58eb\u306e\u6bd4\u306f\u3001\u5e73\u5747\u8cbb\u7528\/\u52b9\u7528\u6bd4\u3068\u547c\u3076\u3002<\/p>\n\n\n\n
    ><\/span> sum(<\/span>est.cost.AA)\/sum(<\/span>est.util.AA)<\/span>\n[1]<\/span> 52.42698<\/span>\n<\/code><\/pre>\n\n\n\n
    C\/F\u7fa4\u306f\u300123.24053\u3060\u3063\u305f\u306e\u306b\u5bfe\u3057\u3001A\u7fa4\u306f\u300152.42698\u3068\u500d\u4ee5\u4e0a\u306e\u8cbb\u7528\u5bfe\u52b9\u7528\u6bd4\u3002<\/p>\n\n\n\n
    \u3088\u3063\u3066\u3001C\/F\u7fa4\u306e\u307b\u3046\u304c\u3001\u8cbb\u7528\u5bfe\u52b9\u7528\u6bd4\u304c\u4f4e\u304f\u6291\u3048\u3089\u308c\u3066\u3044\u308b\u3068\u8a00\u3048\u308b\u3002<\/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>
    \u5224\u65ad\u5206\u6790\u30e2\u30c7\u30eb Decision analysis model \u305d\u306e2\uff08Exercise 6\uff09<\/h2>\n\n\n\n
    \u65b0\u85ac\u3092\u4f7f\u3063\u305f\u5834\u5408\u3068\u5f93\u6765\u85ac\u3092\u4f7f\u3063\u305f\u5834\u5408\u306e\u8cbb\u7528\u3068\u52b9\u679c\u306e\u5897\u5206\u3092\u3001\u526f\u4f5c\u7528\u3042\u308a\u30fb\u306a\u3057\u3068\u3082\u306b\u542b\u3081\u3066\u306e\u30b7\u30ca\u30ea\u30aa\u5168\u4f53\u3067\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
    \u307e\u305a\u3001\u5171\u901a\u306e\u6761\u4ef6\u8a2d\u5b9a\u3002\u4e0b\u75e2\u3084\u809d\u6a5f\u80fd\u969c\u5bb3\u306e\u526f\u4f5c\u7528\u306e\u6761\u4ef6\u3002<\/p>\n\n\n\n
    treat.day <-<\/span> 84<\/span>\ndia.admit <-<\/span> 0.70<\/span>\ndia.cost.o <-<\/span> 150<\/span>\ndia.cost.a <-<\/span> 20000<\/span>\nliv.cost.a <-<\/span> 30000<\/span><\/code><\/pre>\n\n\n\n
    A\uff08\u65b0\u85ac\u3092\u4f7f\u3063\u305f\u5834\u5408\uff09\u3001B\uff08\u5f93\u6765\u306e\u5834\u5408\uff09 2\u3064\u306e\u6761\u4ef6\u3002<\/p>\n\n\n\n
    \u526f\u4f5c\u7528\u304c\u8d77\u304d\u308b\u53ef\u80fd\u6027\u3084\u30b3\u30b9\u30c8\u306e\u30d1\u30bf\u30fc\u30f3\u3002<\/p>\n\n\n\n
    survival <-<\/span> c(0.35,0.20)<\/span>\ndiarrhea <-<\/span> c(0.02,0.10)<\/span>\nliverdys <-<\/span> c(0.15,0.01)<\/span>\ncan.cost <-<\/span> c(2875,2500)<\/span>\nlivertes <-<\/span> c(1500,1000)<\/span>\n<\/code><\/pre>\n\n\n\n
    4\u3064\u306e\u30b7\u30ca\u30ea\u30aa\u3002<\/p>\n\n\n\n
      \n
    1. \u526f\u4f5c\u7528\u306a\u3057<\/li>\n\n\n\n
    2. \u4e0b\u75e2\u306e\u526f\u4f5c\u7528\u3042\u308a\u5916\u6765\u3067\u6cbb\u7642<\/li>\n\n\n\n
    3. \u4e0b\u75e2\u306e\u526f\u4f5c\u7528\u3042\u308a\u5165\u9662\u3057\u3066\u5143\u306e\u6cbb\u7642\u306f\u4e2d\u6b62<\/li>\n\n\n\n
    4. \u809d\u6a5f\u80fd\u969c\u5bb3\u767a\u751f<\/li>\n<\/ol>\n\n\n\n
      \u306e4\u3064\u3002<\/p>\n\n\n\n
      \u305d\u308c\u305e\u308c\u306b\u5408\u308f\u305b\u3066\u30b3\u30b9\u30c8\u306e\u7b97\u51fa\u3002<\/p>\n\n\n\n
      scenario <- c(\"\u526f\u4f5c\u7528\u306a\u3057\",\"\u4e0b\u75e2\u5916\u6765\u52a0\u7642\",\"\u4e0b\u75e2\u5165\u9662\u52a0\u7642\u6cbb\u7642\u4e2d\u6b62\",\"\u809d\u6a5f\u80fd\u969c\u5bb3\")\ntotal.cost <- rbind(\n  (can.cost+livertes)*treat.day,\n  (can.cost+livertes+dia.cost.o)*treat.day,\n  (can.cost+livertes)*14+(dia.cost.o+dia.cost.a)*14,\n  (can.cost+livertes)*treat.day+liv.cost.a*28\n)\ncolnames(total.cost) <- c(\"A\",\"B\")\nrownames(total.cost) <- scenario\ntotal.cost\n<\/code><\/pre>\n\n\n\n
      \u30b3\u30b9\u30c8\u306e\u5408\u8a08\u984d\u3092\u8868\u793a\u3059\u308b\u3068\u3053\u3046\u306a\u308b\u3002<\/p>\n\n\n\n
      > total.cost\n                           A       B\n\u526f\u4f5c\u7528\u306a\u3057            367500  294000\n\u4e0b\u75e2\u5916\u6765\u52a0\u7642          380100  306600\n\u4e0b\u75e2\u5165\u9662\u52a0\u7642\u6cbb\u7642\u4e2d\u6b62  343350  331100\n\u809d\u6a5f\u80fd\u969c\u5bb3           1207500 1134000<\/code><\/pre>\n\n\n\n
      4\u3064\u306e\u30b7\u30ca\u30ea\u30aa\u305d\u308c\u305e\u308c\u304c\u8d77\u304d\u308b\u78ba\u7387\u3092\u8a08\u7b97\u3002<\/p>\n\n\n\n
      probabilit <- rbind(\n  1-diarrhea-liverdys,\n  diarrhea*(1-dia.admit),\n  diarrhea*dia.admit,\n  liverdys\n)\ncolnames(probabilit) <- c(\"A\",\"B\")\nrownames(probabilit) <- scenario\nprobabilit\n<\/code><\/pre>\n\n\n\n
      4\u3064\u306e\u30b7\u30ca\u30ea\u30aa\u304c\u8d77\u304d\u308b\u78ba\u7387\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3002<\/p>\n\n\n\n
      > probabilit\n                         A    B\n\u526f\u4f5c\u7528\u306a\u3057           0.830 0.89\n\u4e0b\u75e2\u5916\u6765\u52a0\u7642         0.006 0.03\n\u4e0b\u75e2\u5165\u9662\u52a0\u7642\u6cbb\u7642\u4e2d\u6b62 0.014 0.07\n\u809d\u6a5f\u80fd\u969c\u5bb3           0.150 0.01<\/code><\/pre>\n\n\n\n
      \u30b3\u30b9\u30c8\u3068\u751f\u3058\u3046\u308b\u78ba\u7387\u3092\u304b\u3051\u308b\u3068\u3001\u671f\u5f85\u8cbb\u7528\u304c\u7b97\u51fa\u3067\u304d\u308b\u3002<\/p>\n\n\n\n
      > total.cost * probabilit\n                            A      B\n\u526f\u4f5c\u7528\u306a\u3057           305025.0 261660\n\u4e0b\u75e2\u5916\u6765\u52a0\u7642           2280.6   9198\n\u4e0b\u75e2\u5165\u9662\u52a0\u7642\u6cbb\u7642\u4e2d\u6b62   4806.9  23177\n\u809d\u6a5f\u80fd\u969c\u5bb3           181125.0  11340<\/code><\/pre>\n\n\n\n
      \u7e26\u65b9\u5411\u306b\u5408\u8a08\u3059\u308b\u3068\u3001\u60a3\u8005\u4e00\u4eba\u5f53\u305f\u308a\u306e\u671f\u5f85\u8cbb\u7528\u304c\u8a08\u7b97\u3067\u304d\u308b\u3002<\/p>\n\n\n\n
      list(\"\u60a3\u80051\u4eba\u5f53\u305f\u308a\u671f\u5f85\u8cbb\u7528\"=colSums(total.cost * probabilit))<\/code><\/pre>\n\n\n\n
      A\u3060\u306850\u4e07\u8fd1\u304f\u3001B\u3060\u306830\u4e07\u3061\u3087\u3063\u3068\u306b\u306a\u308b\u3002<\/p>\n\n\n\n
      > list(\"\u60a3\u80051\u4eba\u5f53\u305f\u308a\u671f\u5f85\u8cbb\u7528\"=colSums(total.cost * probabilit))\n$\u60a3\u80051\u4eba\u5f53\u305f\u308a\u671f\u5f85\u8cbb\u7528\n       A        B \n493237.5 305375.0 <\/code><\/pre>\n\n\n\n
      \u8cbb\u7528\/\u52b9\u679c\u6bd4\u306f\u3001\u60a3\u8005\u4e00\u4eba\u5f53\u305f\u308a\u306e\u671f\u5f85\u8cbb\u7528\u3092\u751f\u5b58\u5272\u5408\u3067\u5272\u308b\u3002<\/p>\n\n\n\n
      > list(\"\u8cbb\u7528\/\u52b9\u679c\u6bd4\"=colSums(total.cost * probabilit)\/survival)\n$`\u8cbb\u7528\/\u52b9\u679c\u6bd4`\n      A       B \n1409250 1526875 <\/code><\/pre>\n\n\n\n
      \u5f93\u6765\u306e\u6cbb\u7642\u304b\u3089\u65b0\u3057\u3044\u6cbb\u7642\u306b\u306a\u3063\u3066\u3001\u3069\u308c\u3060\u3051\u8cbb\u7528\u5bfe\u52b9\u679c\u304c\u5897\u3057\u305f\u304b\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
      \u60a3\u8005\u4e00\u4eba\u5f53\u305f\u308a\u306e\u671f\u5f85\u8cbb\u7528\u3092\u5dee\u5206\u3092\u3001\u751f\u5b58\u5272\u5408\u306e\u5dee\u5206\u3067\u5272\u308a\u7b97\u3059\u308b\u3068\u3001\u5897\u5206\u8cbb\u7528\/\u52b9\u679c\u6bd4\u304c\u8a08\u7b97\u3055\u308c\u308b\u3002<\/p>\n\n\n\n
      inc.cost <- colSums(total.cost * probabilit)\ninc.rati <- (inc.cost[1]-inc.cost[2])\/(survival[1]-survival[2])\nnames(inc.rati) <- \"\u5897\u5206\u8cbb\u7528\/\u52b9\u679c\u6bd4\"\ninc.rati<\/code><\/pre>\n\n\n\n
      \u5897\u5206\u8cbb\u7528\/\u52b9\u679c\u6bd4\u304c\u3069\u308c\u3060\u3051\u306a\u3089\u3001\u610f\u5473\u5408\u3044\u304c\u3042\u308b\u3068\u3059\u308b\u304b\u304c\u3001\u65b0\u85ac\u306e\u610f\u5473\u5408\u3044\u306b\u304b\u304b\u3063\u3066\u304f\u308b\u3002<\/p>\n\n\n\n
      \u4f8b\u984c\u3067\u306f\u3001125\u4e07\u8d85\u3002<\/p>\n\n\n\n
      > inc.rati\n\u5897\u5206\u8cbb\u7528\/\u52b9\u679c\u6bd4 \n        1252417 <\/code><\/pre>\n\n\n\n
      \u30de\u30eb\u30b3\u30d5\u30e2\u30c7\u30eb Markov model \u305d\u306e1\uff08Exercise 8\uff09<\/h2>\n\n\n\n
      \u30de\u30eb\u30b3\u30d5\u30e2\u30c7\u30eb\u3068\u306f\u3001\u73fe\u5728\u306e\u72b6\u614b\u304b\u3089\u306e\u307f\u6b21\u306e\u72b6\u614b\u304c\u767a\u751f\u3059\u308b\u3068\u3044\u3046\u8003\u3048\u306e\u3082\u3068\u3001\u6642\u9593\u3068\u3068\u3082\u306b\u5909\u5316\u3059\u308b\u78ba\u7387\u30e2\u30c7\u30eb\u306e\u3053\u3068\u3067\u3042\u308b\u3002<\/p>\n\n\n\n
      \u3042\u308b\u75c5\u524d\u72b6\u614b\u304c\u3042\u308a\u3001\u6b21\u306e\u4e2d\u7b49\u5ea6\u306e\u75be\u75c5\u306b\u304b\u304b\u308a\u3001\u305d\u306e\u6b21\u306e\u3082\u3063\u3068\u91cd\u7be4\u306a\u75be\u75c5\u306b\u304b\u304b\u308a\u3001\u6700\u7d42\u7684\u306b\u6b7b\u3092\u8fce\u3048\u308b\u3088\u3046\u306a\u751f\u6d3b\u7fd2\u6163\u75c5\u306e\u8003\u3048\u65b9\u306b\u3042\u3066\u306f\u3081\u3084\u3059\u3044\u30e2\u30c7\u30eb\u3068\u3044\u3048\u308b\u3002<\/p>\n\n\n\n
      \u3053\u3053\u3067\u306f\u3001\u9ad8\u8102\u8840\u75c7\u306e\u60a3\u8005\u3055\u3093\u304c\u3001\u72ed\u5fc3\u75c7\u306b\u304b\u304b\u308a\u3001\u5fc3\u7b4b\u6897\u585e\u306b\u304b\u304b\u308a\u3001\u6b7b\u3092\u8fce\u3048\u308b\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3092\u884c\u3046\u3002<\/p>\n\n\n\n
      S1 <-<\/span> c(0.90,0.00,0.00,0.00)<\/span>\nS2 <-<\/span> c(0.06,0.83,0.88,0.00)<\/span>\nS3 <-<\/span> c(0.04,0.12,0.05,0.00)<\/span>\nS4 <-<\/span> c(0.00,0.05,0.07,1.00)<\/span>\n<\/code><\/pre>\n\n\n\n
      \u4e0a\u8a18\u6761\u4ef6\u3067\u3001\u30de\u30eb\u30b3\u30d5\u9023\u9396\u8a08\u7b97\u3092\u3059\u308b\u3002<\/p>\n\n\n\n
      my.markov <- function(n){\n  nu <- c(100,0,0,0)\n  dat.nu <- nu\n  S <- matrix(c(S1,S2,S3,S4),nr=4)\n  for (i in 1:n){\n    nu <- nu%*%S\n    dat.nu <- rbind(dat.nu,nu)\n  }\n  dat.nu\n}\n<\/code><\/pre>\n\n\n\n
      \u30de\u30eb\u30b3\u30d5\u9023\u9396\u309220\u56de\u8a08\u7b97\u3057\u3066\u307f\u308b\u3002<\/p>\n\n\n\n
      dat <- my.markov(20)\ncolnames(dat) <- c(\"\u9ad8\u8102\u8840\u75c7\",\"\u72ed\u5fc3\u75c7\",\"\u5fc3\u7b4b\u6897\u585e\",\"\u6b7b\")\ndat<\/code><\/pre>\n\n\n\n
      \u3069\u306e\u72b6\u614b\u304c\u3069\u306e\u3088\u3046\u306b\u5897\u3048\u3066\u3044\u304f\u306e\u304b\u306e\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u3002<\/p>\n\n\n\n
      > dat\n        \u9ad8\u8102\u8840\u75c7   \u72ed\u5fc3\u75c7 \u5fc3\u7b4b\u6897\u585e        \u6b7b\ndat.nu 100.00000  0.00000 0.000000  0.000000\n        90.00000  6.00000 4.000000  0.000000\n        81.00000 13.90000 4.520000  0.580000\n        72.90000 20.37460 5.134000  1.591400\n        65.61000 25.80284 5.617652  2.969510\n        59.04900 30.29649 6.001623  4.652888\n        53.14410 33.97045 6.297620  6.587826\n        47.82969 36.92603 6.517100  8.727182\n        43.04672 39.25343 6.670166 11.029680\n        38.74205 41.03290 6.765789 13.459263\n        34.86784 42.33572 6.811919 15.984514\n        31.38106 43.22521 6.815597 18.578134\n        28.24295 43.75751 6.783047 21.216486\n        25.41866 43.98239 6.719772 23.879175\n        22.87679 43.94391 6.630622 26.548679\n        20.58911 43.68100 6.519872 29.210018\n        18.53020 43.22806 6.391278 31.850459\n        16.67718 42.61543 6.248139 34.459251\n        15.00946 41.86980 6.093346 37.027392\n        13.50852 41.01464 5.929422 39.547417\n        12.15767 40.07056 5.758569 42.013208<\/code><\/pre>\n\n\n\n
      \u56f3\u306b\u3057\u3066\u307f\u308b\u3068\u3053\u3046\u306a\u308b\u3002<\/p>\n\n\n\n
      matplot(<\/span>dat)<\/span>\n<\/code><\/pre>\n\n\n\n
      \"\"<\/figure>\n\n\n\n
      <\/span><\/p>\n\n\n\n
      \u7a4d\u307f\u4e0a\u3052\u8a08\u7b97\u3092\u3057\u305f\u30b0\u30e9\u30d5\u306b\u3059\u308b\u3068\u3001<\/p>\n\n\n\n
      x1 <-<\/span> dat[,1]<\/span>\nx2 <-<\/span> x1+<\/span>dat[,2]<\/span>\nx3 <-<\/span> x2+<\/span>dat[,3]<\/span>\nx4 <-<\/span> x3+<\/span>dat[,4]<\/span>\nmatplot(cbind(<\/span>x1,<\/span>x2,<\/span>x3,<\/span>x4))<\/span>\n<\/code><\/pre>\n\n\n\n
      \u3053\u3046\u306a\u308b\u3002<\/p>\n\n\n\n
      \"\"<\/figure>\n\n\n\n
      <\/span><\/p>\n\n\n\n
      \u30de\u30eb\u30b3\u30d5\u30e2\u30c7\u30eb Markov model \u305d\u306e2\uff08Exercise 9\uff09<\/h2>\n\n\n\n
      \u6761\u4ef6\u304c2\u3064\u3042\u3063\u305f\u5834\u5408\u306e\u6bd4\u8f03\u3092\u884c\u3063\u3066\u307f\u308b\u3002Sa\u3068Sb\u306e2\u3064\u306e\u6761\u4ef6\u304c\u3042\u3063\u305f\u3068\u3059\u308b\u3002<\/p>\n\n\n\n
      Sa1 <- c(0.90,0.00,0.00,0.00)\nSa2 <- c(0.06,0.83,0.88,0.00)\nSa3 <- c(0.04,0.12,0.05,0.00)\nSa4 <- c(0.00,0.05,0.07,1.00)\nSb1 <- c(0.93,0.00,0.00,0.00)\nSb2 <- c(0.05,0.85,0.90,0.00)\nSb3 <- c(0.02,0.10,0.03,0.00)\nSb4 <- c(0.00,0.05,0.07,1.00)<\/code><\/pre>\n\n\n\n
      2 \u3064\u306e\u30de\u30eb\u30b3\u30d5\u9023\u9396\u30e2\u30c7\u30eb\u306e\u8a08\u7b97\u30b9\u30af\u30ea\u30d7\u30c8\u3092\u4f5c\u6210\u3059\u308b\u3002<\/p>\n\n\n\n
      my.markovA <- function(n){\n  nu <- c(100,0,0,0)\n  dat.nu <- nu\n  S <- matrix(c(Sa1,Sa2,Sa3,Sa4),nr=4)\n  for (i in 1:n){\n    nu <- nu%*%S\n    dat.nu <- rbind(dat.nu,nu)\n  }\n  dat.nu\n}\n\nmy.markovB <- function(n){\n  nu <- c(100,0,0,0)\n  dat.nu <- nu\n  S <- matrix(c(Sb1,Sb2,Sb3,Sb4),nr=4)\n  for (i in 1:n){\n    nu <- nu%*%S\n    dat.nu <- rbind(dat.nu,nu)\n  }\n  dat.nu\n}<\/code><\/pre>\n\n\n\n
      2\u3064\u306e\u6761\u4ef6\u306e\u30de\u30eb\u30b3\u30d5\u9023\u9396\u89e3\u6790\u3092\u884c\u3046\u3002<\/p>\n\n\n\n
      \u65b0\u305f\u306a\u6b7b\u4ea1\u3060\u3051\u3092\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u306b\u6b8b\u3059\u3002<\/p>\n\n\n\n
      datA <- my.markovA(20)\ncolnames(datA) <- c(\"\u9ad8\u8102\u8840\u75c7\",\"\u72ed\u5fc3\u75c7\",\"\u5fc3\u7b4b\u6897\u585e\",\"\u6b7b\")\nlast.yr.deathA <- c(0,datA[1:20,4])\ndatB <- my.markovB(20)\ncolnames(datB) <- c(\"\u9ad8\u8102\u8840\u75c7\",\"\u72ed\u5fc3\u75c7\",\"\u5fc3\u7b4b\u6897\u585e\",\"\u6b7b\")\nlast.yr.deathB <- c(0,datB[1:20,4])\ndatA1 <- data.frame(datA,last.yr.deathA)\ndatA1$\u65b0\u305f\u306a\u6b7b <- datA1$\u6b7b - datA1$last.yr.deathA\ndatB1 <- data.frame(datB,last.yr.deathB)\ndatB1$\u65b0\u305f\u306a\u6b7b <- datB1$\u6b7b - datB1$last.yr.deathB\ndatA1 <- datA1[,c(1,2,3,6)]\ndatB1 <- datB1[,c(1,2,3,6)]<\/code><\/pre>\n\n\n\n
      A\u7fa4\u306e\u30c7\u30fc\u30bf\u306f\u4ee5\u4e0b\u306e\u3088\u3046\u306b\u7b97\u51fa\u3055\u308c\u305f\u3002<\/p>\n\n\n\n
      > datA1\n        \u9ad8\u8102\u8840\u75c7   \u72ed\u5fc3\u75c7 \u5fc3\u7b4b\u6897\u585e \u65b0\u305f\u306a\u6b7b\ndat.nu 100.00000  0.00000 0.000000 0.000000\nX       90.00000  6.00000 4.000000 0.000000\nX.1     81.00000 13.90000 4.520000 0.580000\nX.2     72.90000 20.37460 5.134000 1.011400\nX.3     65.61000 25.80284 5.617652 1.378110\nX.4     59.04900 30.29649 6.001623 1.683378\nX.5     53.14410 33.97045 6.297620 1.934938\nX.6     47.82969 36.92603 6.517100 2.139356\nX.7     43.04672 39.25343 6.670166 2.302498\nX.8     38.74205 41.03290 6.765789 2.429583\nX.9     34.86784 42.33572 6.811919 2.525250\nX.10    31.38106 43.22521 6.815597 2.593621\nX.11    28.24295 43.75751 6.783047 2.638352\nX.12    25.41866 43.98239 6.719772 2.662689\nX.13    22.87679 43.94391 6.630622 2.669504\nX.14    20.58911 43.68100 6.519872 2.661339\nX.15    18.53020 43.22806 6.391278 2.640441\nX.16    16.67718 42.61543 6.248139 2.608793\nX.17    15.00946 41.86980 6.093346 2.568141\nX.18    13.50852 41.01464 5.929422 2.520024\nX.19    12.15767 40.07056 5.758569 2.465792<\/code><\/pre>\n\n\n\n
      B\u7fa4\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3002<\/p>\n\n\n\n
      > datB1\n        \u9ad8\u8102\u8840\u75c7   \u72ed\u5fc3\u75c7 \u5fc3\u7b4b\u6897\u585e \u65b0\u305f\u306a\u6b7b\ndat.nu 100.00000  0.00000 0.000000 0.000000\nX       93.00000  5.00000 2.000000 0.000000\nX.1     86.49000 10.70000 2.420000 0.390000\nX.2     80.43570 15.59750 2.872400 0.704400\nX.3     74.80520 19.86482 3.254636 0.980943\nX.4     69.56884 23.55453 3.580225 1.221066\nX.5     64.69902 26.72199 3.854236 1.428342\nX.6     60.17009 29.41746 4.081807 1.605896\nX.7     55.95818 31.68697 4.267602 1.756599\nX.8     52.04111 33.57268 4.415889 1.883081\nX.9     48.39823 35.11313 4.530566 1.987746\nX.10    45.01035 36.34358 4.615195 2.072796\nX.11    41.85963 37.29624 4.673021 2.140243\nX.12    38.92946 38.00050 4.707007 2.191923\nX.13    36.20439 38.48321 4.719850 2.229516\nX.14    33.67009 38.76881 4.714004 2.254550\nX.15    31.31318 38.87960 4.691703 2.268421\nX.16    29.12126 38.83585 4.654974 2.272399\nX.17    27.08277 38.65601 4.605659 2.267641\nX.18    25.18698 38.35684 4.545426 2.255197\nX.19    23.42389 37.95355 4.475786 2.236022<\/code><\/pre>\n\n\n\n
      \u30b3\u30b9\u30c8\u3092\u5b9a\u7fa9\u3059\u308b\u3002<\/p>\n\n\n\n
      cost <- c(100,300,750,2500)\nnames(cost) <- c(\"\u9ad8\u8102\u8840\u75c7\",\"\u72ed\u5fc3\u75c7\",\"\u5fc3\u7b4b\u6897\u585e\",\"\u6b7b\u4ea1\")<\/code><\/pre>\n\n\n\n
      \u305d\u308c\u305e\u308c\u306e\u75c5\u614b\u306b\u30b3\u30b9\u30c8\u3092\u639b\u3051\u5408\u308f\u305b\u308b\u3002<\/p>\n\n\n\n
      datA2 <- cbind(\"\u9ad8\u8102\u8840\u75c7\"=datA1[,1]*cost[1],\n               \"\u72ed\u5fc3\u75c7\"=datA1[,2]*cost[2],\n               \"\u5fc3\u7b4b\u6897\u585e\"=datA1[,3]*cost[3],\n               \"\u65b0\u305f\u306a\u6b7b\u4ea1\"=datA1[,4]*cost[4])\ndatB2 <- cbind(\"\u9ad8\u8102\u8840\u75c7\"=datB1[,1]*cost[1],\n               \"\u72ed\u5fc3\u75c7\"=datB1[,2]*cost[2],\n               \"\u5fc3\u7b4b\u6897\u585e\"=datB1[,3]*cost[3],\n               \"\u65b0\u305f\u306a\u6b7b\u4ea1\"=datB1[,4]*cost[4])<\/code><\/pre>\n\n\n\n
      A\u7fa4\u3068B\u7fa4\u3001\u305d\u308c\u305e\u308c\u306e\u30b3\u30b9\u30c8\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
      > datA2\n       \u9ad8\u8102\u8840\u75c7    \u72ed\u5fc3\u75c7 \u5fc3\u7b4b\u6897\u585e \u65b0\u305f\u306a\u6b7b\u4ea1\n [1,] 10000.000     0.000    0.000      0.000\n [2,]  9000.000  1800.000 3000.000      0.000\n [3,]  8100.000  4170.000 3390.000   1450.000\n [4,]  7290.000  6112.380 3850.500   2528.500\n [5,]  6561.000  7740.851 4213.239   3445.275\n [6,]  5904.900  9088.947 4501.217   4208.444\n [7,]  5314.410 10191.136 4723.215   4837.345\n [8,]  4782.969 11077.809 4887.825   5348.390\n [9,]  4304.672 11776.030 5002.625   5756.246\n[10,]  3874.205 12309.870 5074.342   6073.958\n[11,]  3486.784 12700.717 5108.939   6313.125\n[12,]  3138.106 12967.563 5111.697   6484.051\n[13,]  2824.295 13127.254 5087.286   6595.881\n[14,]  2541.866 13194.718 5039.829   6656.722\n[15,]  2287.679 13183.172 4972.967   6673.759\n[16,]  2058.911 13104.299 4889.904   6653.347\n[17,]  1853.020 12968.418 4793.458   6601.102\n[18,]  1667.718 12784.628 4686.105   6521.981\n[19,]  1500.946 12560.940 4570.009   6420.353\n[20,]  1350.852 12304.393 4447.066   6300.060\n[21,]  1215.767 12021.167 4318.927   6164.479<\/code><\/pre>\n\n\n\n
      \u5e74\u3042\u305f\u308a\u306e\u5408\u8a08\u30b3\u30b9\u30c8\u3068\u300120\u5e74\u9593\u306e\u30c8\u30fc\u30bf\u30eb\u30b3\u30b9\u30c8\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
      sum.year.costA <-<\/span> matrix(c(<\/span>datA1[,1],<\/span>datA1[,2],<\/span>datA1[,3],<\/span>datA1[,4]),<\/span>nc=4)%*%matrix(<\/span>cost,<\/span>nr=4)<\/span>\nsum.tota.costA <-<\/span> sum(<\/span>sum.year.costA)<\/span>\nsum.year.costB <-<\/span> matrix(c(<\/span>datB1[,1],<\/span>datB1[,2],<\/span>datB1[,3],<\/span>datB1[,4]),<\/span>nc=4)%*%matrix(<\/span>cost,<\/span>nr=4)<\/span>\nsum.tota.costB <-<\/span> sum(<\/span>sum.year.costB)<\/span>\n<\/code><\/pre>\n\n\n\n
      \u30b3\u30b9\u30c8\u306fB\u306e\u307b\u3046\u304cA\u3088\u308a\u308259000\u307b\u3069\u5b89\u304f\u306a\u308b\u3002<\/p>\n\n\n\n
      > sum.tota.costA\n[1] 500944.6\n> sum.tota.costB\n[1] 442204.8\n> sum.tota.costA - sum.tota.costB\n[1] 58739.81<\/code><\/pre>\n\n\n\n
      \u30de\u30eb\u30b3\u30d5\u30e2\u30c7\u30eb\u3067\u4e88\u6e2c\u3057\u305fA\u7fa4\u3068B\u7fa4\u3092\u7a4d\u307f\u4e0a\u3052\u30b0\u30e9\u30d5\u3067\u63cf\u753b\u3057\u3066\u307f\u308b\u3068\u3001\u4ee5\u4e0b\u306e\u3068\u304a\u308a\u306e\u30b0\u30e9\u30d5\u306b\u306a\u308b\u3002<\/p>\n\n\n\n
      B\u7fa4\u306e\u307b\u3046\u304c\u9ad8\u8102\u8840\u75c7\u3067\u3068\u3069\u307e\u308b\u4eba\u304c\u591a\u3044\u3002<\/p>\n\n\n\n
      barplot(t(datA),col=c(4:1))\nlegend(3,30,c(\"\u6b7b\",\"\u5fc3\u7b4b\u6897\u585e\",\"\u72ed\u5fc3\u75c7\",\"\u9ad8\u8102\u8840\u75c7\"),bg=\"white\",fill=c(1:4))\nbarplot(t(datB),col=c(4:1))\nlegend(3,30,c(\"\u6b7b\",\"\u5fc3\u7b4b\u6897\u585e\",\"\u72ed\u5fc3\u75c7\",\"\u9ad8\u8102\u8840\u75c7\"),bg=\"white\",fill=c(1:4))\n<\/code><\/pre>\n\n\n\n