{"id":331,"date":"2018-09-25T22:12:32","date_gmt":"2018-09-25T13:12:32","guid":{"rendered":"https:\/\/best-biostatistics.com\/toukei-er\/entry\/principal-component-regression-and-partial-least-squares-regression-in-r\/"},"modified":"2024-10-12T18:10:25","modified_gmt":"2024-10-12T09:10:25","slug":"principal-component-regression-and-partial-least-squares-regression-in-r","status":"publish","type":"post","link":"https:\/\/best-biostatistics.com\/toukei-er\/entry\/principal-component-regression-and-partial-least-squares-regression-in-r\/","title":{"rendered":"R \u3067\u4e3b\u6210\u5206\u56de\u5e30\u3068\u90e8\u5206\u7684\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\u3092\u5b9f\u884c\u3059\u308b\u65b9\u6cd5"},"content":{"rendered":"\n

\u4e3b\u6210\u5206\u56de\u5e30\u3068\u90e8\u5206\u7684\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\u3092 R \u3067\u5b9f\u884c\u3059\u308b\u65b9\u6cd5\u306e\u89e3\u8aac<\/p>\n\n\n\n\n\n\n\n

\u90e8\u5206\u7684\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\u3068\u306f<\/h2>\n\n\n\n

\u90e8\u5206\u7684\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\u306e\u524d\u306b\u3001\u4e3b\u6210\u5206\u56de\u5e30\u3092\u8aac\u660e\u3059\u308b\u3002<\/p>\n\n\n\n

\u4e3b\u6210\u5206\u56de\u5e30\uff08Principal Component Regression, PCR\uff09\u306f\u3001\u4e3b\u6210\u5206\u5206\u6790\u3068\u56de\u5e30\u5206\u6790\u306e\u878d\u5408\u3002<\/p>\n\n\n\n

\u4e3b\u6210\u5206\u5206\u6790\u3067\u60c5\u5831\u306e\u96c6\u7d04\u3092\u3057\u3066\u3001\u5909\u6570\u3092\u6e1b\u3089\u3057\u3066\u304b\u3089\u56de\u5e30\u5206\u6790\u3092\u884c\u3046\u65b9\u6cd5\u3002<\/p>\n\n\n\n

\u591a\u91cd\u5171\u7dda\u6027\u304c\u5fc3\u914d\u306a\u5909\u6570\u304c\u542b\u307e\u308c\u3066\u3044\u3066\u3082\u3001\u4e3b\u6210\u5206\u5f97\u70b9\u306b\u96c6\u7d04\u3055\u308c\u308b\u305f\u3081\u554f\u984c\u304c\u306a\u304f\u306a\u308b\u3002<\/p>\n\n\n\n

\u90e8\u5206\u7684\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\uff08Partial Least Squares Regression, PLS Regression\uff09\u306f\u3001\u4e3b\u6210\u5206\u56de\u5e30\u306e\u767a\u5c55\u7248\u3002<\/p>\n\n\n\n

\u72ec\u7acb\u5909\u6570\u306f\u4e3b\u6210\u5206\u56de\u5e30\u3068\u540c\u3058\u3002<\/p>\n\n\n\n

PLS\u306f\u72ec\u7acb\u5909\u6570\u3060\u3051\u3067\u306a\u304f\u3001\u5f93\u5c5e\u5909\u6570\u306b\u3082\u65b0\u305f\u306a\u5f97\u70b9\u3092\u60f3\u5b9a\u3059\u308b\u3002<\/p>\n\n\n\n

\u72ec\u7acb\u5909\u6570\u3082\u5f93\u5c5e\u5909\u6570\u3082\u9055\u3046\u4e16\u754c\u306b\u6295\u5c04\u3057\u3066\u6f5c\u5728\u7684\u306a\u95a2\u4fc2\u3092\u898b\u308b\u3002<\/p>\n\n\n\n

\u90e8\u5206\u7684\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\u306e\u5206\u6790\u4f8b\u306e\u305f\u3081\u306b\u30c7\u30fc\u30bf\u3068\u30d1\u30c3\u30b1\u30fc\u30b8\u306e\u6e96\u5099<\/h2>\n\n\n\n

\u30c7\u30fc\u30bf\u306f\u3001ISLR\u30d1\u30c3\u30b1\u30fc\u30b8\u306eCollege\u30c7\u30fc\u30bf\u3092\u7528\u3044\u308b\u3002<\/p>\n\n\n\n

\u7c73\u56fd\u306e777\u5927\u5b66\u306e\u51fa\u9858\u8005\u3084\u5b66\u751f\u6570\u3001\u66f8\u7c4d\u8cfc\u5165\u306b\u304b\u304b\u308b\u30b3\u30b9\u30c8\u306a\u3069\u3068\u3044\u304618\u500b\u306e\u30d7\u30ed\u30d5\u30a1\u30a4\u30eb\u30c7\u30fc\u30bf\u3002<\/p>\n\n\n\n

\u89e3\u6790\u30d1\u30c3\u30b1\u30fc\u30b8\u306fpls\u30d1\u30c3\u30b1\u30fc\u30b8\u3002<\/p>\n\n\n\n

\u4e3b\u6210\u5206\u56de\u5e30\uff08PCR\uff09\u3082\u90e8\u5206\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\uff08PLS\uff09\u3082\u3067\u304d\u308b\u3002<\/p>\n\n\n\n

\u6700\u521d\u306e\u4e00\u56de\u3060\u3051\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3059\u308b\u3002<\/p>\n\n\n\n

install.packages<\/span>(<\/span>\"ISLR\"<\/span>)<\/span>\ninstall.packages<\/span>(<\/span>\"pls\"<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\u4f7f\u3046\u3068\u304d\u306flibrary()\u3067\u547c\u3073\u51fa\u3059\u3002<\/p>\n\n\n\n
library<\/span>(<\/span>ISLR)<\/span>\nlibrary<\/span>(<\/span>pls)<\/span>\n<\/code><\/pre>\n\n\n\n
\u4f8b\u984c\u306f\u3001\u767b\u9332\u3059\u308b\u5b66\u751f\u6570\uff08Enroll\uff09\u3092\u4e88\u6e2c\u3059\u308b\u30e2\u30c7\u30eb\u3092\u4f5c\u308b\u3053\u3068\u3002<\/p>\n\n\n\n
\u307e\u305a\u3001\u8aac\u660e\u5909\u6570\u540c\u58eb\u306e\u76f8\u95a2\u4fc2\u6570\u3092\u898b\u3066\u307f\u308b\u3002<\/p>\n\n\n\n
Apps\uff08\u51fa\u9858\u8005\u6570\uff09\u3068Accept\uff08\u51fa\u9858\u8005\u53d7\u4ed8\u4eba\u6570\uff09\u306e\u3088\u3046\u306b0.9\u3092\u8d85\u3048\u308b\u76f8\u95a2\u4fc2\u6570\u3082\u898b\u3089\u308c\u308b\u3002<\/p>\n\n\n\n
><\/span> round<\/span>(<\/span>cor<\/span>(<\/span>College[,<\/span>c<\/span>(<\/span>-<\/span>1<\/span>,<\/span>-<\/span>4<\/span>)]),<\/span>2<\/span>)<\/span>\nApps Accept Top10perc Top25perc F.Undergrad P.Undergrad Outstate Room.Board Books Personal   PhD Terminal S.F.Ratio perc.alumni Expend Grad.Rate\nApps         1.00<\/span>   0.94<\/span>      0.34<\/span>      0.35<\/span>        0.81<\/span>        0.40<\/span>     0.05<\/span>       0.16<\/span>  0.13<\/span>     0.18<\/span>  0.39<\/span>     0.37<\/span>      0.10<\/span>       -<\/span>0.09<\/span>   0.26<\/span>      0.15<\/span>\nAccept       0.94<\/span>   1.00<\/span>      0.19<\/span>      0.25<\/span>        0.87<\/span>        0.44<\/span>    -<\/span>0.03<\/span>       0.09<\/span>  0.11<\/span>     0.20<\/span>  0.36<\/span>     0.34<\/span>      0.18<\/span>       -<\/span>0.16<\/span>   0.12<\/span>      0.07<\/span>\nTop10perc    0.34<\/span>   0.19<\/span>      1.00<\/span>      0.89<\/span>        0.14<\/span>       -<\/span>0.11<\/span>     0.56<\/span>       0.37<\/span>  0.12<\/span>    -<\/span>0.09<\/span>  0.53<\/span>     0.49<\/span>     -<\/span>0.38<\/span>        0.46<\/span>   0.66<\/span>      0.49<\/span>\nTop25perc    0.35<\/span>   0.25<\/span>      0.89<\/span>      1.00<\/span>        0.20<\/span>       -<\/span>0.05<\/span>     0.49<\/span>       0.33<\/span>  0.12<\/span>    -<\/span>0.08<\/span>  0.55<\/span>     0.52<\/span>     -<\/span>0.29<\/span>        0.42<\/span>   0.53<\/span>      0.48<\/span>\nF.Undergrad  0.81<\/span>   0.87<\/span>      0.14<\/span>      0.20<\/span>        1.00<\/span>        0.57<\/span>    -<\/span>0.22<\/span>      -<\/span>0.07<\/span>  0.12<\/span>     0.32<\/span>  0.32<\/span>     0.30<\/span>      0.28<\/span>       -<\/span>0.23<\/span>   0.02<\/span>     -<\/span>0.08<\/span>\nP.Undergrad  0.40<\/span>   0.44<\/span>     -<\/span>0.11<\/span>     -<\/span>0.05<\/span>        0.57<\/span>        1.00<\/span>    -<\/span>0.25<\/span>      -<\/span>0.06<\/span>  0.08<\/span>     0.32<\/span>  0.15<\/span>     0.14<\/span>      0.23<\/span>       -<\/span>0.28<\/span>  -<\/span>0.08<\/span>     -<\/span>0.26<\/span>\nOutstate     0.05<\/span>  -<\/span>0.03<\/span>      0.56<\/span>      0.49<\/span>       -<\/span>0.22<\/span>       -<\/span>0.25<\/span>     1.00<\/span>       0.65<\/span>  0.04<\/span>    -<\/span>0.30<\/span>  0.38<\/span>     0.41<\/span>     -<\/span>0.55<\/span>        0.57<\/span>   0.67<\/span>      0.57<\/span>\nRoom.Board   0.16<\/span>   0.09<\/span>      0.37<\/span>      0.33<\/span>       -<\/span>0.07<\/span>       -<\/span>0.06<\/span>     0.65<\/span>       1.00<\/span>  0.13<\/span>    -<\/span>0.20<\/span>  0.33<\/span>     0.37<\/span>     -<\/span>0.36<\/span>        0.27<\/span>   0.50<\/span>      0.42<\/span>\nBooks        0.13<\/span>   0.11<\/span>      0.12<\/span>      0.12<\/span>        0.12<\/span>        0.08<\/span>     0.04<\/span>       0.13<\/span>  1.00<\/span>     0.18<\/span>  0.03<\/span>     0.10<\/span>     -<\/span>0.03<\/span>       -<\/span>0.04<\/span>   0.11<\/span>      0.00<\/span>\nPersonal     0.18<\/span>   0.20<\/span>     -<\/span>0.09<\/span>     -<\/span>0.08<\/span>        0.32<\/span>        0.32<\/span>    -<\/span>0.30<\/span>      -<\/span>0.20<\/span>  0.18<\/span>     1.00<\/span> -<\/span>0.01<\/span>    -<\/span>0.03<\/span>      0.14<\/span>       -<\/span>0.29<\/span>  -<\/span>0.10<\/span>     -<\/span>0.27<\/span>\nPhD          0.39<\/span>   0.36<\/span>      0.53<\/span>      0.55<\/span>        0.32<\/span>        0.15<\/span>     0.38<\/span>       0.33<\/span>  0.03<\/span>    -<\/span>0.01<\/span>  1.00<\/span>     0.85<\/span>     -<\/span>0.13<\/span>        0.25<\/span>   0.43<\/span>      0.31<\/span>\nTerminal     0.37<\/span>   0.34<\/span>      0.49<\/span>      0.52<\/span>        0.30<\/span>        0.14<\/span>     0.41<\/span>       0.37<\/span>  0.10<\/span>    -<\/span>0.03<\/span>  0.85<\/span>     1.00<\/span>     -<\/span>0.16<\/span>        0.27<\/span>   0.44<\/span>      0.29<\/span>\nS.F.Ratio    0.10<\/span>   0.18<\/span>     -<\/span>0.38<\/span>     -<\/span>0.29<\/span>        0.28<\/span>        0.23<\/span>    -<\/span>0.55<\/span>      -<\/span>0.36<\/span> -<\/span>0.03<\/span>     0.14<\/span> -<\/span>0.13<\/span>    -<\/span>0.16<\/span>      1.00<\/span>       -<\/span>0.40<\/span>  -<\/span>0.58<\/span>     -<\/span>0.31<\/span>\nperc.alumni -<\/span>0.09<\/span>  -<\/span>0.16<\/span>      0.46<\/span>      0.42<\/span>       -<\/span>0.23<\/span>       -<\/span>0.28<\/span>     0.57<\/span>       0.27<\/span> -<\/span>0.04<\/span>    -<\/span>0.29<\/span>  0.25<\/span>     0.27<\/span>     -<\/span>0.40<\/span>        1.00<\/span>   0.42<\/span>      0.49<\/span>\nExpend       0.26<\/span>   0.12<\/span>      0.66<\/span>      0.53<\/span>        0.02<\/span>       -<\/span>0.08<\/span>     0.67<\/span>       0.50<\/span>  0.11<\/span>    -<\/span>0.10<\/span>  0.43<\/span>     0.44<\/span>     -<\/span>0.58<\/span>        0.42<\/span>   1.00<\/span>      0.39<\/span>\nGrad.Rate    0.15<\/span>   0.07<\/span>      0.49<\/span>      0.48<\/span>       -<\/span>0.08<\/span>       -<\/span>0.26<\/span>     0.57<\/span>       0.42<\/span>  0.00<\/span>    -<\/span>0.27<\/span>  0.31<\/span>     0.29<\/span>     -<\/span>0.31<\/span>        0.49<\/span>   0.39<\/span>      1.00<\/span>\n<\/code><\/pre>\n\n\n\n
Enroll\u3092\u4ed6\u306e\u3059\u3079\u3066\u306e\u5909\u6570\u3067\u4e88\u6e2c\u3059\u308b\u91cd\u56de\u5e30\u30e2\u30c7\u30eb\u306b\u3042\u3066\u306f\u3081\u3066\u307f\u3066\u3001VIF\u3092\u8a08\u7b97\u3057\u3066\u307f\u308b\u3002<\/p>\n\n\n\n
VIF\u306b\u3064\u3044\u3066\u306f\u4ee5\u4e0b\u3092\u53c2\u7167\u3002<\/p>\n\n\n
\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\u3042\u308f\u305b\u3066\u8aad\u307f\u305f\u3044<\/span>\n\t\t\t\t\t
\"\"<\/figure><\/div>\t\t\t\t\t
\n\t\t\t\t\t\tR \u3067\u591a\u91cd\u5171\u7dda\u6027\u3092\u30c1\u30a7\u30c3\u30af\u3059\u308b\u305f\u3081\u306b VIF \u3092\u8a08\u7b97\u3059\u308b\u65b9\u6cd5<\/a>\n\t\t\t\t\t\t\u91cd\u56de\u5e30\u5206\u6790\u3067\u3001\u5909\u6570\u9078\u629e\u306b\u304a\u3044\u3066\u3001\u8003\u616e\u3057\u306a\u3044\u3068\u3044\u3051\u306a\u3044\u306e\u304c\u3001\u591a\u91cd\u5171\u7dda\u6027\u3002 \u591a\u91cd\u5171\u7dda\u6027\u306e\u30c1\u30a7\u30c3\u30af\u306f\u7d71\u8a08\u30bd\u30d5\u30c8R\u3067\u3069\u3046\u3084\u308b\u306e\u304b\uff1f \u591a\u91cd\u5171\u7dda\u6027\u3068\u306f\u4f55\u304b\uff1f \u751f\u307e\u308c\u305f\u897f\u66a6\u5e74…<\/span>\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t<\/div>\n\n\n\n\n\n\n
\u6700\u521d\u306e\u5909\u6570 (Private\u3068\u3044\u3046\u5909\u6570) \u3092\u524a\u9664\u3057\u3066\u3001Enroll\u3092\u76ee\u7684\u5909\u6570\u306b\u3001\u305d\u306e\u4ed6\u306e\u5909\u6570\u3092\u8aac\u660e\u5909\u6570\u306b\u3057\u3066\u91cd\u56de\u5e30\u5206\u6790\u3092\u884c\u3044\u3001car \u30d1\u30c3\u30b1\u30fc\u30b8\u306e vif() \u95a2\u6570\u3067VIF\u3092\u8a08\u7b97\u3055\u305b\u305f\u3002<\/p>\n\n\n\n
\u305d\u306e\u7d50\u679c\u3001Apps\u3068Accept\u304c10\u3092\u8d85\u3048\u308b\u6570\u5024\u3068\u306a\u308a\u3001\u76f8\u95a2\u4fc2\u65700.94\u3092\u8003\u616e\u3059\u308b\u3068\u3001\u3053\u308c\u3089\u304c\u591a\u91cd\u5171\u7dda\u6027\u3092\u751f\u3058\u3066\u3044\u308b\u3068\u8a00\u3048\u308b\u3002<\/p>\n\n\n\n
><\/span> College1 <-<\/span> College[,<\/span> -<\/span>1<\/span>]<\/span>\n><\/span> lm.res1 <-<\/span> lm<\/span>(<\/span>Enroll ~<\/span> .,<\/span> data=<\/span>College1)<\/span>\n><\/span> libarary<\/span>(<\/span>car)<\/span>\n><\/span> round<\/span>(<\/span>vif<\/span>(<\/span>lm.res1),<\/span>1<\/span>)<\/span>\nApps      Accept   Top10perc   Top25perc F.Undergrad P.Undergrad    Outstate  Room.Board       Books    Personal         PhD    Terminal   S.F.Ratio perc.alumni      Expend   Grad.Rate\n13.5<\/span>        17.2<\/span>         7.5<\/span>         5.6<\/span>         6.7<\/span>         1.7<\/span>         3.9<\/span>         2.0<\/span>         1.1<\/span>         1.3<\/span>         4.1<\/span>         4.0<\/span>         1.9<\/span>         1.8<\/span>         3.1<\/span>         1.8<\/span>\n<\/code><\/pre>\n\n\n\n
\u4e3b\u6210\u5206\u56de\u5e30\u5206\u6790\u306b\u5165\u308b\u524d\u306b\u3001\u5b66\u7fd2\u30c7\u30fc\u30bf\u3068\u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u306b\u5206\u3051\u308b\u3002<\/p>\n\n\n\n
5\u5206\u306e4\u3092\u5b66\u7fd2\u30c7\u30fc\u30bf\u30015\u5206\u306e1\u3092\u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u306b\u3059\u308b\u3002<\/p>\n\n\n\n
set.seed<\/span>(<\/span>20180924<\/span>)<\/span>\nsub <-<\/span> sample<\/span>(<\/span>1<\/span>:<\/span>777<\/span>,<\/span> round<\/span>(<\/span>777<\/span>*<\/span>4<\/span>\/<\/span>5<\/span>))<\/span>\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>
\u90e8\u5206\u7684\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\u306e\u524d\u306b \u4e3b\u6210\u5206\u56de\u5e30\u306e\u5206\u6790\u4f8b<\/h2>\n\n\n\n
\u4e3b\u6210\u5206\u56de\u5e30\u3092 R \u3067\u5b9f\u884c\u3057\u3066\u307f\u308b\u3002<\/p>\n\n\n\n
pls\u30d1\u30c3\u30b1\u30fc\u30b8\u306epcr()\u3092\u4f7f\u3046\u3002<\/p>\n\n\n\n
\u30af\u30ed\u30b9\u30d0\u30ea\u30c7\u30fc\u30b7\u30e7\u30f3\u3092\u884c\u3044\u306a\u304c\u3089\u30e2\u30c7\u30eb\u3092\u4f5c\u308b\u305f\u3081\u306b\u3001validation=”CV”\u3092\u6307\u5b9a\u3059\u308b\u3002<\/p>\n\n\n\n
\u5b66\u7fd2\u30c7\u30fc\u30bf\u3067\u3001\u30e2\u30c7\u30eb\u3092\u4f5c\u308b\u3002<\/p>\n\n\n\n
pcr.res.cv <-<\/span> pcr<\/span>(<\/span>Enroll ~<\/span> .,<\/span> data=<\/span>College[<\/span>sub,],<\/span> validation=<\/span>\"CV\"<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\u7d50\u679c\u306e\u30b5\u30de\u30ea\u30fc\u3092\u78ba\u8a8d\u3002<\/p>\n\n\n\n
Enroll\u3092\u9664\u304d\u5168\u90e8\u306717\u5909\u6570\u306a\u306e\u3067\u300117\u306e\u5408\u6210\u5909\u6570\u304c\u4f5c\u6210\u3055\u308c\u308b\u3002<\/p>\n\n\n\n
><\/span> summary<\/span>(<\/span>pcr.res.cv)<\/span>\nData:<\/span>   X dimension:<\/span> 622<\/span> 17<\/span>\nY dimension:<\/span> 622<\/span> 1<\/span>\nFit method:<\/span> svdpc\nNumber of components considered:<\/span> 17<\/span>\nVALIDATION:<\/span> RMSEP\nCross-<\/span>validated using 10<\/span> random segments.\n(<\/span>Intercept)<\/span>  1<\/span> comps  2<\/span> comps  3<\/span> comps  4<\/span> comps  5<\/span> comps  6<\/span> comps  7<\/span> comps  8<\/span> comps\nCV           868.6<\/span>    791.8<\/span>    217.1<\/span>    207.8<\/span>    207.5<\/span>    191.5<\/span>    192.6<\/span>    192.9<\/span>    188.5<\/span>\nadjCV        868.6<\/span>    806.7<\/span>    216.9<\/span>    207.6<\/span>    207.3<\/span>    191.2<\/span>    192.2<\/span>    192.6<\/span>    188.0<\/span>\n9<\/span> comps  10<\/span> comps  11<\/span> comps  12<\/span> comps  13<\/span> comps  14<\/span> comps  15<\/span> comps  16<\/span> comps\nCV       188.7<\/span>     188.9<\/span>     187.9<\/span>     188.4<\/span>     186.4<\/span>     186.5<\/span>     187.1<\/span>     187.6<\/span>\nadjCV    188.1<\/span>     188.3<\/span>     187.3<\/span>     187.8<\/span>     185.8<\/span>     185.9<\/span>     186.4<\/span>     186.9<\/span>\n17<\/span> comps\nCV        188.0<\/span>\nadjCV     187.3<\/span>\nTRAINING:<\/span> % variance explained\n1<\/span> comps  2<\/span> comps  3<\/span> comps  4<\/span> comps  5<\/span> comps  6<\/span> comps  7<\/span> comps  8<\/span> comps  9<\/span> comps\nX         47.34<\/span>    87.44<\/span>    94.47<\/span>    96.84<\/span>    98.42<\/span>    99.15<\/span>    99.62<\/span>    99.97<\/span>   100.00<\/span>\nEnroll    20.93<\/span>    93.90<\/span>    94.39<\/span>    94.50<\/span>    95.40<\/span>    95.43<\/span>    95.43<\/span>    95.76<\/span>    95.76<\/span>\n10<\/span> comps  11<\/span> comps  12<\/span> comps  13<\/span> comps  14<\/span> comps  15<\/span> comps  16<\/span> comps  17<\/span> comps\nX         100.00<\/span>    100.00<\/span>    100.00<\/span>    100.00<\/span>    100.00<\/span>    100.00<\/span>    100.00<\/span>    100.00<\/span>\nEnroll     95.76<\/span>     95.81<\/span>     95.82<\/span>     95.89<\/span>     95.89<\/span>     95.91<\/span>     95.91<\/span>     95.91<\/span>\n<\/code><\/pre>\n\n\n\n
\u4e3b\u6210\u5206\u6570\u3054\u3068\u306eMean Squared Error of Prediction (MSEP)\u3092\u78ba\u8a8d\u3002<\/p>\n\n\n\n
plot<\/span>(<\/span>MSEP<\/span>(<\/span>pcr.res.cv),<\/span> legendpos=<\/span>\"topright\"<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\u7b2c\u4e8c\u4e3b\u6210\u5206\u3067MSEP\u306f\u6fc0\u6e1b\u3057\u3066\u305d\u308c\u4ee5\u964d\u306f\u5927\u304d\u304f\u306f\u5909\u308f\u3089\u306a\u3044\u3002<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
\u8a08\u7b97\u4e0a\u306e\u6700\u9069\u306a\u4e3b\u6210\u5206\u306e\u6570\u3092\u78ba\u8a8d\u3002<\/p>\n\n\n\n
\u7b54\u3048\u306f5\u3002<\/p>\n\n\n\n
selectNcomp<\/span>(<\/span>pcr.res.cv,<\/span> method=<\/span>\"onesigma\"<\/span>,<\/span> plot=<\/span>T<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
\u4e3b\u6210\u5206\u304c5\u306e\u6642\u306eMSEP\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
\u4e3b\u6210\u5206\u304c2\u306e\u6642\u306eMSEP\u3082\u8a08\u7b97\u3057\u3066\u307f\u308b\u3002<\/p>\n\n\n\n
MSEP<\/span>(<\/span>pcr.res.cv,<\/span> ncomp=<\/span>5<\/span>)<\/span>\nMSEP<\/span>(<\/span>pcr.res.cv,<\/span> ncomp=<\/span>2<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
5\u306e\u307b\u3046\u304cMSEP\u304c\u5c0f\u3055\u3044\u3053\u3068\u306f\u5c0f\u3055\u3044\u3002<\/p>\n\n\n\n
><\/span> MSEP<\/span>(<\/span>pcr.res.cv,<\/span> ncomp=<\/span>5<\/span>)<\/span>\n(<\/span>Intercept)<\/span>  5<\/span> comps\nCV          754513<\/span>    36689<\/span>\nadjCV       754513<\/span>    36571<\/span>\n><\/span> MSEP<\/span>(<\/span>pcr.res.cv,<\/span> ncomp=<\/span>2<\/span>)<\/span>\n(<\/span>Intercept)<\/span>  2<\/span> comps\nCV          754513<\/span>    47134<\/span>\nadjCV       754513<\/span>    47046<\/span>\n<\/code><\/pre>\n\n\n\n
\u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u306b\u9069\u7528\u3057\u3066\u3001MSEP\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
MSEP\u306f\u5b66\u7fd2\u30c7\u30fc\u30bf\u306b\u6bd4\u3079\u3066\u3060\u3044\u3076\u5927\u304d\u304f\u306a\u3063\u305f\u3002<\/p>\n\n\n\n
\u4e88\u6e2c\u6027\u80fd\u306f\u9ad8\u304f\u306a\u3044\u3002<\/p>\n\n\n\n
><\/span> MSEP<\/span>(<\/span>pcr.res.cv,<\/span> ncomp=<\/span>5<\/span>,<\/span> newdata=<\/span>College[<\/span>-<\/span>sub,])<\/span>\n(<\/span>Intercept)<\/span>      5<\/span> comps\n1317758<\/span>        95394<\/span>\n<\/code><\/pre>\n\n\n\n
\u90e8\u5206\u7684\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\u306e\u5206\u6790\u4f8b<\/h2>\n\n\n\n
\u90e8\u5206\u7684\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\uff08Partial Least Squares\uff09\u306f\u3001\u540c\u3058\u982d\u6587\u5b57\u8a9e\u306b\u306a\u308b\u3001Projection to Latent Structure\u3068\u3082\u547c\u3070\u308c\u308b\u3002<\/p>\n\n\n\n
Projection to Latent Structure\uff08\u6f5c\u5728\u7684\u69cb\u9020\u3078\u306e\u6295\u5c04\uff09\u306e\u307b\u3046\u304c\u3001\u8a08\u7b97\u65b9\u6cd5\u3092\u53cd\u6620\u3057\u305f\u540d\u524d\u3060\u3002<\/p>\n\n\n\n
\u4f7f\u3046\u95a2\u6570\u306fplsr()\u3067\u3001\u66f8\u304d\u65b9\u306fpcr()\u3068\u540c\u3058\u3002<\/p>\n\n\n\n
plsr.res.cv <-<\/span> plsr<\/span>(<\/span>Enroll ~<\/span> .,<\/span> data=<\/span>College[<\/span>sub,],<\/span> validation=<\/span>\"CV\"<\/span>)<\/span>\nsummary<\/span>(<\/span>plsr.res.cv)<\/span>\n<\/code><\/pre>\n\n\n\n
\u4e3b\u6210\u5206\u56de\u5e30\u3068\u6bd4\u8f03\u3057\u3066\u3001\u7b2c\u4e00\u4e3b\u6210\u5206\u304b\u3089 Enroll \u306e \uff05Variance \u304c\u9ad8\u3044\u3002<\/p>\n\n\n\n
\u5f53\u3066\u306f\u307e\u308a\u304c\u3044\u3044\u3053\u3068\u304c\u671f\u5f85\u3055\u308c\u308b\u3002<\/p>\n\n\n\n
><\/span> summary<\/span>(<\/span>plsr.res.cv)<\/span>\nData:<\/span>   X dimension:<\/span> 622<\/span> 17<\/span>\nY dimension:<\/span> 622<\/span> 1<\/span>\nFit method:<\/span> kernelpls\nNumber of components considered:<\/span> 17<\/span>\nVALIDATION:<\/span> RMSEP\nCross-<\/span>validated using 10<\/span> random segments.\n(<\/span>Intercept)<\/span>  1<\/span> comps  2<\/span> comps  3<\/span> comps  4<\/span> comps  5<\/span> comps  6<\/span> comps  7<\/span> comps  8<\/span> comps\nCV           868.6<\/span>    227.6<\/span>    216.8<\/span>    203.8<\/span>    196.6<\/span>    198.2<\/span>    195.9<\/span>    195.0<\/span>    195.4<\/span>\nadjCV        868.6<\/span>    227.1<\/span>    216.3<\/span>    203.3<\/span>    195.9<\/span>    197.2<\/span>    194.9<\/span>    194.1<\/span>    194.5<\/span>\n9<\/span> comps  10<\/span> comps  11<\/span> comps  12<\/span> comps  13<\/span> comps  14<\/span> comps  15<\/span> comps  16<\/span> comps\nCV       195.3<\/span>     195.1<\/span>     193.8<\/span>     194.4<\/span>     195.1<\/span>     195.3<\/span>     195.5<\/span>     195.6<\/span>\nadjCV    194.4<\/span>     194.0<\/span>     192.8<\/span>     193.4<\/span>     194.0<\/span>     194.2<\/span>     194.4<\/span>     194.5<\/span>\n17<\/span> comps\nCV        196.3<\/span>\nadjCV     195.2<\/span>\nTRAINING:<\/span> % variance explained\n1<\/span> comps  2<\/span> comps  3<\/span> comps  4<\/span> comps  5<\/span> comps  6<\/span> comps  7<\/span> comps  8<\/span> comps  9<\/span> comps\nX         42.16<\/span>    86.69<\/span>    93.55<\/span>    96.13<\/span>    97.72<\/span>    98.80<\/span>    99.49<\/span>    99.85<\/span>   100.00<\/span>\nEnroll    93.65<\/span>    94.32<\/span>    95.05<\/span>    95.57<\/span>    95.68<\/span>    95.76<\/span>    95.76<\/span>    95.76<\/span>    95.76<\/span>\n10<\/span> comps  11<\/span> comps  12<\/span> comps  13<\/span> comps  14<\/span> comps  15<\/span> comps  16<\/span> comps  17<\/span> comps\nX         100.00<\/span>    100.00<\/span>     100.0<\/span>    100.00<\/span>    100.00<\/span>    100.00<\/span>    100.00<\/span>    100.00<\/span>\nEnroll     95.87<\/span>     95.89<\/span>      95.9<\/span>     95.91<\/span>     95.91<\/span>     95.91<\/span>     95.91<\/span>     95.91<\/span>\n<\/code><\/pre>\n\n\n\n
\u7b2c\u4e00\u4e3b\u6210\u5206\u3060\u3051\u3067MSEP\u306f\u30ac\u30af\u30f3\u3068\u6e1b\u5c11\u3059\u308b\u3002<\/p>\n\n\n\n
\u305d\u306e\u3042\u3068\u306f\u306a\u3060\u3089\u304b\u306b\u6e1b\u5c11\u3002<\/p>\n\n\n\n
plot<\/span>(<\/span>MSEP<\/span>(<\/span>plsr.res.cv),<\/span> legendpos=<\/span>\"topright\"<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
\u6700\u9069\u306a\u4e3b\u6210\u5206\u6570\u306f4\u3068\u793a\u3055\u308c\u305f\u3002<\/p>\n\n\n\n
selectNcomp<\/span>(<\/span>plsr.res.cv,<\/span> method=<\/span>\"onesigma\"<\/span>,<\/span> plot=<\/span>T<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
\u4e3b\u6210\u5206\u65704\u306e\u3068\u304d\u306eMSEP\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
\u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u306eMSEP\u3068\u6bd4\u8f03\u3059\u308b\u3002<\/p>\n\n\n\n
MSEP<\/span>(<\/span>plsr.res.cv,<\/span> ncomp=<\/span>4<\/span>)<\/span>\nMSEP<\/span>(<\/span>plsr.res.cv,<\/span> ncomp=<\/span>4<\/span>,<\/span> newdata=<\/span>College[<\/span>-<\/span>sub,])<\/span>\n<\/code><\/pre>\n\n\n\n
\u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u306eMSEP\u306f\u3050\u3093\u3068\u5897\u3048\u3066\u3057\u307e\u3063\u3066\u5f53\u3066\u306f\u307e\u308a\u306f\u3088\u304f\u306a\u304b\u3063\u305f\u3002<\/p>\n\n\n\n
><\/span> MSEP<\/span>(<\/span>plsr.res.cv,<\/span> ncomp=<\/span>4<\/span>)<\/span>\n(<\/span>Intercept)<\/span>  4<\/span> comps\nCV          754513<\/span>    38670<\/span>\nadjCV       754513<\/span>    38384<\/span>\n><\/span> MSEP<\/span>(<\/span>plsr.res.cv,<\/span> ncomp=<\/span>4<\/span>,<\/span> newdata=<\/span>College[<\/span>-<\/span>sub,])<\/span>\n(<\/span>Intercept)<\/span>      4<\/span> comps\n1317758<\/span>        89507<\/span>\n<\/code><\/pre>\n\n\n\n
\u307e\u3068\u3081<\/h2>\n\n\n\n
\u72ec\u7acb\u5909\u6570\u306b\u591a\u91cd\u5171\u7dda\u6027\u304c\u7591\u308f\u308c\u308b\u3068\u304d\u306b\u6709\u5229\u306a\u4e3b\u6210\u5206\u56de\u5e30\u3068\u90e8\u5206\u7684\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\u3002<\/p>\n\n\n\n
\u4eca\u56de\u306f\u305f\u307e\u305f\u307e\u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u3078\u306e\u5f53\u3066\u306f\u307e\u308a\u304c\u60aa\u304b\u3063\u305f\u3060\u3051\u3068\u601d\u3046\u3002<\/p>\n\n\n\n
\u666e\u901a\u306e\u6700\u5c0f\u4e8c\u4e57\u6cd5\u306a\u3089\u3069\u3061\u3089\u304b\u3092\u3084\u3081\u306a\u3051\u308c\u3070\u306a\u3089\u306a\u3044\u3088\u3046\u306a\u3001\u3068\u3066\u3082\u76f8\u95a2\u304c\u5f37\u304f\u3066\u3001\u591a\u91cd\u5171\u7dda\u6027\u304c\u7591\u308f\u308c\u308b\u5909\u6570\u3092\u540c\u6642\u306b\u6271\u3044\u305f\u3044\u5834\u5408\u306f\u3001\u3069\u3061\u3089\u306e\u5909\u6570\u3082\u3084\u3081\u306a\u304f\u3066\u3044\u3044\u4e3b\u6210\u5206\u56de\u5e30\u3084\u90e8\u5206\u7684\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\u3092\u304a\u8a66\u3057\u3042\u308c\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"
\u4e3b\u6210\u5206\u56de\u5e30\u3068\u90e8\u5206\u7684\u6700\u5c0f\u4e8c\u4e57\u56de\u5e30\u3092 R \u3067\u5b9f\u884c\u3059\u308b\u65b9\u6cd5\u306e\u89e3\u8aac<\/p>\n","protected":false},"author":2,"featured_media":2721,"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":false,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[5,102,83,103],"tags":[],"class_list":["post-331","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-r","category-102","category-83","category-103"],"jetpack_publicize_connections":[],"jetpack_featured_media_url":"https:\/\/best-biostatistics.com\/toukei-er\/wp-content\/uploads\/2018\/09\/20180924181828.png","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/posts\/331","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=331"}],"version-history":[{"count":2,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/posts\/331\/revisions"}],"predecessor-version":[{"id":2724,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/posts\/331\/revisions\/2724"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/media\/2721"}],"wp:attachment":[{"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/media?parent=331"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/categories?post=331"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/best-biostatistics.com\/toukei-er\/wp-json\/wp\/v2\/tags?post=331"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}