{"id":1181,"date":"2019-05-03T00:55:56","date_gmt":"2019-05-02T15:55:56","guid":{"rendered":"http:\/\/141.164.34.82\/?p=1181"},"modified":"2019-05-03T05:13:43","modified_gmt":"2019-05-02T20:13:43","slug":"1%eb%b3%80%ec%88%98-%ec%9a%94%ec%95%bd%ed%86%b5%ea%b3%84%eb%9f%89","status":"publish","type":"post","link":"http:\/\/ds.sumeun.org\/?p=1181","title":{"rendered":"1\ubcc0\uc218 \uc694\uc57d\ud1b5\uacc4\ub7c9"},"content":{"rendered":"<h1>1\ubcc0\uc218 \uae30\uc220 \ud1b5\uacc4\ub7c9<\/h1>\n<p>\uae30\uc220 \ud1b5\uacc4\ub7c9(Descriptive Statitistics)\uc740 \uc8fc\uc5b4\uc9c4 \uc790\ub8cc\ub97c \uba87 \uac1c\uc758 \uc22b\uc790\ub85c \uc694\uc57d\ud558\uc5ec \ub098\ud0c0\ub0b4\uae30 \uc704\ud574 \uad6c\ud55c\ub2e4. \uc608\ub97c \ub4e4\uc5b4 \ud3c9\uade0\uc774\ub098 \ubd84\uc0b0, \ud45c\uc900\ud3b8\ucc28 \ub4f1\uc744 \uad6c\ud558\ub294 \uac83\uc740 \uc790\ub8cc\uc758 \uc8fc\uc694 \ud2b9\uc131\uc778 \uc911\uc2ec\uc758 \uc704\uce58\uc640 \ud37c\uc9d0 \uc815\ub3c4\ub97c \ub4dc\ub7ec\ub0b8\ub2e4.<\/p>\n<p>\uc694\uc57d \ud1b5\uacc4\uce58\ub294 \ud06c\uac8c <strong>\uc9d1\uc911\uacbd\ud5a5\uce58<\/strong>\uc640 <strong>\ubcc0\uc0b0\uc131 \uce21\uc815\uce58<\/strong>\ub85c \uad6c\ubd84\ud560 \uc218 \uc788\ub2e4. <\/p>\n<ul>\n<li>\n<p><strong>\uc9d1\uc911\uacbd\ud5a5\uce58<\/strong> : \uc790\ub8cc\uc758 \uc911\uc2ec\uc774 \uc5b4\ub514\uc778\uc9c0\ub97c \ub098\ud0c0\ub0b8\ub2e4. <strong>\uc911\uc2ec<\/strong>\uc744 \uc5b4\ub5bb\uac8c \uc815\uc758\ud558\ub290\ub0d0\uc5d0 \ub530\ub77c \ud1b5\uacc4\ub7c9\uc774 \ub2ec\ub77c\uc9c4\ub2e4. \ud3c9\uade0(mean), \uc911\uc559\uac12(median), \ucd5c\ube48\uac12(mode), \uc808\uc0ac\ud3c9\uade0(trimmed mean) \ub4f1\uc774 \ub9ce\uc774 \uc4f0\uc778\ub2e4.<\/p>\n<\/li>\n<li>\n<p><strong>\ubcc0\uc0b0\uc131 \uce21\uc815\uce58<\/strong> : \uc790\ub8cc\uac00 \uc5bc\ub9c8\ub098 \ub113\uac8c \ubd84\ud3ec\ub418\uc5b4 \uc788\ub294\uac00\ub97c \ub098\ud0c0\ub0b8\ub2e4. \ubc94\uc704(range), \uc0ac\ubd84\uc810\uac04 \ubc94\uc704(inter-quartile range), \ubd84\uc0b0(variance), \ud45c\uc900\ud3b8\ucc28(standard deviation) \ub4f1\uc774 \ub9ce\uc774 \uc4f0\uc778\ub2e4.<\/p>\n<\/li>\n<li>\n<p>\uadf8 \uc678\uc758 \uc694\uc57d \ud1b5\uacc4\uce58 : \uadf8 \uc678\uc5d0\ub3c4 \ubd84\ud3ec\uc758 \ub300\uce6d\uc131\uc744 \ub4dc\ub7ec\ub0b4\ub294 <strong>\uc65c\ub3c4(skewness; \ub300\uce6d\ub3c4)<\/strong>\uc640 \ubd84\ud3ec\uc758 \ubfb0\uc871\ud55c \uc815\ub3c4\ub97c \ub098\ud0c0\ub0b4\ub294 <strong>\ucca8\ub3c4(kurtosis)<\/strong>\uac00 \uc788\ub2e4. <\/p>\n<\/li>\n<\/ul>\n<h2>R \uad6c\ud604: \uc9d1\uc911\uacbd\ud5a5\uce58<\/h2>\n<ul>\n<li>\ud3c9\uade0 : <code>mean()<\/code><\/li>\n<li>\uc911\uc559\uac12 : <code>median()<\/code><\/li>\n<li>\ucd5c\ube48\uac12 : <code>Mode()<\/code>. \uc544\ub798\uc758 \ud568\uc218\ub294 <a href=\"https:\/\/stackoverflow.com\/questions\/2547402\/is-there-a-built-in-function-for-finding-the-mode\">stackoverflow<\/a>\uc5d0\uc11c \uac00\uc838\uc654\ub2e4.<\/li>\n<\/ul>\n<pre><code class=\"r\">Mode &lt;- function(x) {\n  ux &lt;- unique(x)\n  ux[which.max(tabulate(match(x, ux)))]\n}\n<\/code><\/pre>\n<ul>\n<li>\uc808\uc0ac\ud3c9\uade0 : <code>mean( , trim = 0.2)<\/code>. <code>trim=<\/code> : \uc790\ub8cc\uc758 \uc591\uadf9\ub2e8\uc5d0\uc11c \uc81c\uc678(\uc808\uc0ac)\ub418\ub294 \ube44\uc728<\/li>\n<\/ul>\n<pre><code class=\"r\">x &lt;- rnorm(10)\nx &lt;- c(x, 2, 2)\n\n# \uc9d1\uc911\uacbd\ud5a5\uce58\nmean(x)\n<\/code><\/pre>\n<pre>## [1] 0.3060493\n<\/pre>\n<pre><code class=\"r\">median(x)\n<\/code><\/pre>\n<pre>## [1] -0.000349418\n<\/pre>\n<pre><code class=\"r\">Mode(x) # table\uc744 \ud1b5\ud574 \ube48\ub3c4\uc218\ub97c \ud655\uc778\ud560 \uc218 \uc788\ub2e4. \n<\/code><\/pre>\n<pre>## [1] 2\n<\/pre>\n<pre><code class=\"r\">mean(x, trim = 0.2)\n<\/code><\/pre>\n<pre>## [1] 0.072794\n<\/pre>\n<h2>\ubcc0\uc0b0\uc131 \uce21\uc815\uce58<\/h2>\n<ul>\n<li>\ubc94\uc704 : <code>max()-min()<\/code> (\ucd5c\ub300\uac12\uacfc \ucd5c\uc18c\uac12\uc758 \ucc28\uc774)<\/li>\n<li>\ucd5c\ub300\uac12 : <code>max()<\/code><\/li>\n<li>\ucd5c\uc18c\uac12 : <code>min()<\/code><\/li>\n<li>\uc0ac\ubd84\uc810\uac04 \ubc94\uc704 : <code>IQR()<\/code><\/li>\n<li>\ubd84\uc0b0 : <code>var()<\/code><\/li>\n<li>\ud45c\uc900\ud3b8\ucc28 : <code>sd()<\/code><\/li>\n<\/ul>\n<pre><code class=\"r\">x\n<\/code><\/pre>\n<pre>##  [1]  0.22031100 -0.21220588 -0.25715180 -0.01407646  0.72880944\n##  [6] -0.03316143 -0.65260856  0.03771608  0.01337762 -0.15841839\n## [11]  2.00000000  2.00000000\n<\/pre>\n<pre><code class=\"r\"># \ubcc0\uc0b0\uc131 \uce21\uc815\uce58\nmax(x)-min(x)\n<\/code><\/pre>\n<pre>## [1] 2.652609\n<\/pre>\n<pre><code class=\"r\">IQR(x)\n<\/code><\/pre>\n<pre>## [1] 0.5193009\n<\/pre>\n<pre><code class=\"r\">var(x)\n<\/code><\/pre>\n<pre>## [1] 0.7291598\n<\/pre>\n<pre><code class=\"r\">sd(x)\n<\/code><\/pre>\n<pre>## [1] 0.8539086\n<\/pre>\n<h2>\ubc31\ubd84\uc704\uc218<\/h2>\n<p>\ucc38\uace0\ub85c \ubc31\ubd84\uc704\uc218\ub294 \uc911\uc559\uac12, \ucd5c\ub300\uac12, \ucd5c\uc18c\uac12 \ub4f1\uc744 \ubaa8\ub450 \ub098\ud0c0\ub0bc \uc218 \uc788\ub294 \uc77c\ubc18\ud654\ub41c \uac1c\ub150\uc774\ub2e4. <\/p>\n<p>R\uc5d0\uc11c \ubc31\ubd84\uc704\uc218\ub294 <code>quantile()<\/code> \ud568\uc218\ub85c \uad6c\ud55c\ub2e4.<\/p>\n<ul>\n<li>\uc911\uc559\uac12 : 50%-\ubc31\ubd84\uc704\uc218 <code>quantile(x, probs=.5)<\/code><\/li>\n<li>\ucd5c\uc18c\uac12 : 0%-\ubc31\ubd84\uc704\uc218 <code>quantile(x, probs=.0)<\/code><\/li>\n<li>\ucd5c\ub300\uac12 : 100%-\ubc31\ubd84\uc704\uc218 <code>quantile(x, probs=1)<\/code><\/li>\n<li>\uc774 \ubaa8\ub450\ub97c \ud568\uaed8 \uad6c\ud558\ub824\uba74 <code>quantile(x, probs=c(0, 0.5, 1))<\/code><\/li>\n<\/ul>\n<h2>\ubc94\uc8fc\ud615, \uc21c\uc704\ud615 \ubcc0\uc218<\/h2>\n<p>\ub9cc\uc57d \ubc94\uc8fc\ud615 \ub610\ub294 \uc21c\uc704\ud615 \ubcc0\uc218\uc5d0 \ub300\ud574 \uc694\uc57d \ud1b5\uacc4\uce58\ub97c \uad6c\ud558\uace0\uc790 \ud55c\ub2e4\uba74 \uc55e\uc5d0\uc11c \uc598\uae30\ud55c \ub9ce\uc740 \ud1b5\uacc4\ub7c9\uc774 \uc4f8\ubaa8 \uc5c6\ub2e4. \ubc94\uc8fc\ud615\uc744 \uc608\ub85c \ub4e4\uc5b4\ubcf4\uc790. \ub9cc\uc57d \uc11c\uc6b8, \ubd80\uc0b0, \ub300\uad6c\ub97c <code>1<\/code>, <code>2<\/code>, <code>3<\/code>\uc73c\ub85c \ucf54\ub529\uc744 \ud558\uace0, \ube48\ub3c4\uc218\uac00 2,1,3\uc77c \ub54c \ud3c9\uade0\uc744 \uad6c\ud558\uba74 2.1666667\uc774\ub2e4. \ud558\uc9c0\ub9cc \uc774\uac8c \ubb34\uc2a8 \uc758\ubbf8\uac00 \uc788\uc744\uae4c? <\/p>\n<p>\uc11c\uc6b8, \ubd80\uc0b0, \ub300\uad6c\ub97c <code>1<\/code>,<code>2<\/code>,<code>3<\/code>\uc73c\ub85c \ucf54\ub529\ud55c \uac83\uc740 \ub2e8\uc21c\ud788 \ud3b8\uc758\ub97c \uc704\ud574\uc11c\uc774\ub2e4. \uc11c\uc6b8, \ubd80\uc0b0, \ub300\uad6c\ub97c <code>3<\/code>, <code>2<\/code>, <code>1<\/code>\ub85c \ucf54\ub529\ud558\uac70\ub098 <code>2<\/code>, <code>1<\/code>, <code>3<\/code>\uc73c\ub85c \ucf54\ub529\ud574\ub3c4 \ubb34\ubc29\ud558\ub2e4. \ud558\uc9c0\ub9cc \ucf54\ub529 \ubc29\ubc95\uc5d0 \ub530\ub77c \ud3c9\uade0\uc740 \ub2ec\ub77c\uc9c4\ub2e4! \ubd84\uc0b0\uc774\ub098 \ud45c\uc900\ud3b8\ucc28\uc758 \uacbd\uc6b0\ub3c4 \ud3c9\uade0\uacfc \ub9c8\ucc2c\uac00\uc9c0\ub85c \ucf54\ub529 \ubc29\ubc95\uc5d0 \ub530\ub77c \ub2ec\ub77c\uc9c4\ub2e4. \ubc94\uc8fc\ud615 \ubcc0\uc218\uc5d0\uc11c \uc0ac\uc6a9\ud560 \uc218 \uc788\ub294 \ubcc0\uc0b0\uc131 \uce21\uc815\uce58\ub294 <strong>\uc5d4\ud2b8\ub85c\ud53c(entropy)<\/strong>\uc640 <strong>\uc9c0\ub2c8\ubd88\uc21c\ub3c4(gini impurity index)<\/strong>\uac00 \uc788\ub2e4.<\/p>\n<ul>\n<li>\uc5d4\ud2b8\ub85c\ud53c[<sup>1]<\/sup> : <code>entropy::entropy(table(x))<\/code><\/li>\n<li>\uc9c0\ub2c8\ubd88\uc21c\ub3c4[<sup>2]<\/sup> : <code>1-sum(prop.table(table(x))^2)<\/code><\/li>\n<\/ul>\n<pre><code class=\"r\">gini = function(x, ...) { #useNA = &#39;always&#39;, &#39;no&#39;, &#39;ifany&#39;\n  sum(1-prop.table(table(x, ...))^2)\n}\n<\/code><\/pre>\n<p>\ubc94\uc8fc\ud615, \uc21c\uc704\ud615 \uc790\ub8cc\ub97c \uc694\uc57d\ud558\ub294 \ud55c \uac00\uc9c0 \ubc29\ubc95\uc740 \ud45c(\ud14c\uc774\ube14)\uc744 \uc791\uc131\ud558\ub294 \uac83\uc774\ub2e4. <\/p>\n<ul>\n<li><strong>\ube48\ub3c4\ud45c<\/strong> : <code>table()<\/code><\/li>\n<li><strong>\ube44\uc728\ud45c<\/strong> : <code>prop.table()<\/code><\/li>\n<\/ul>\n<p>\ube48\ub3c4\ud45c(<code>table()<\/code>)\ub294 \ub2e4\uc74c\uc758 \uc120\ud0dd\uc0ac\ud56d\uc774 \uc788\ub2e4.<\/p>\n<ul>\n<li><code>NA<\/code>\ub97c \ud3ec\ud568\ud560 \uac83\uc778\uac00? \ud3ec\ud568\ud55c\ub2e4\uba74 \uc5b8\uc81c \ud3ec\ud568\ud560 \uac83\uc778\uac00?\n<ul>\n<li><code>table(x, useNA = &#39;no&#39;)<\/code> : <code>NA<\/code>\ub97c \uc81c\uc678\ud55c \ube48\ub3c4\ud45c<\/li>\n<li><code>table(x, useNA = &#39;ifany&#39;)<\/code> : <code>NA<\/code>\uac00 \uc788\ub2e4\uba74 \ud3ec\ud568\ud574\ub77c<\/li>\n<li><code>table(x, useNA = &#39;always&#39;)<\/code> : <code>NA<\/code>\ub97c \uc5b8\uc81c\ub098 \ud3ec\ud568\ud574\ub77c<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<p><code>prop.table()<\/code>\ub294 \ube48\ub3c4\ud45c\ub97c \uc0ac\uc6a9\ud574\uc11c \ube44\uc728\ud45c\ub97c \ub9cc\ub4e4\uc5b4 \uc900\ub2e4.<\/p>\n<pre><code class=\"r\">prop.table(table(x))\n<\/code><\/pre>\n<p>[<sup>1]:<\/sup> <strong>\uc5d4\ud2b8\ub85c\ud53c<\/strong>\ub294 \ubb34\uc791\uc704\ub3c4\ub77c\uace0\ub3c4 \ubd88\ub9ac\uba70, \\(- \\sum_i p_i \\log(p_i)\\) \ub85c \uad6c\ud55c\ub2e4. \ud558\ub098\uc758 \ubc94\uc8fc\uc758 \ube44\uc728\uc774 \ub192\uc544\uc9c0\uba74 \uc5d4\ud2b8\ub85c\ud53c\ub294 \uc791\uc544\uc9c0\uba70, \ud55c \ubc94\uc8fc\uac00 \ube44\uc728 1\uc77c \ub54c \uc5d4\ud2b8\ub85c\ud53c\ub294 0\uc774 \ub41c\ub2e4. \ucd5c\ub300\uac12\uc740 \\(\\log n\\) ( \\(n\\) : \ubc94\uc8fc\uc758 \uac2f\uc218)\uc774\ub2e4.<\/p>\n<p>[<sup>2]:<\/sup> <strong>\uc9c0\ub2c8\ubd88\uc21c\ub3c4<\/strong>\ub294 \\(1-\\sum_i p_i^2\\) \uc73c\ub85c \uc815\uc758\ub41c\ub2e4. \ube44\uc728\uc5d0 \ub9de\ucdb0\uc11c \uc784\uc758\ub85c, \uadf8\ub9ac\uace0 \ub3c5\ub9bd\uc801\uc73c\ub85c \ubc94\uc8fc\ub97c \uc120\ud0dd\ud588\uc744 \ub54c, \ub450 \ubc94\uc8fc\uac00 \ub2e4\ub97c \ud655\ub960\uc774\ub2e4. \ub9cc\uc57d \ud558\ub098\uc758 \ubc94\uc8fc\uac00 \ud070 \ube44\uc728\uc744 \ucc28\uc9c0\ud55c\ub2e4\uba74 \uc9c0\ub2c8\ubd88\uc21c\ub3c4\ub294 0\uc774 \ub41c\ub2e4. \ub9cc\uc57d \uc5ec\ub7ec \ubc94\uc8fc\uac00 \ube44\uc2b7\ud55c \ube44\uc728\uc744 \ucc28\uc9c0\ud55c\ub2e4\uba74 \uc9c0\ub2c8\ubd88\uc21c\ub3c4\ub294 \ucee4\uc9c4\ub2e4.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>1\ubcc0\uc218 \uae30\uc220 \ud1b5\uacc4\ub7c9 \uae30\uc220 \ud1b5\uacc4\ub7c9(Descriptive Statitistics)\uc740 \uc8fc\uc5b4\uc9c4 \uc790\ub8cc\ub97c \uba87 \uac1c\uc758 \uc22b\uc790\ub85c \uc694\uc57d\ud558\uc5ec \ub098\ud0c0\ub0b4\uae30 \uc704\ud574 \uad6c\ud55c\ub2e4. \uc608\ub97c \ub4e4\uc5b4 \ud3c9\uade0\uc774\ub098 \ubd84\uc0b0, \ud45c\uc900\ud3b8\ucc28 \ub4f1\uc744 \uad6c\ud558\ub294 \uac83\uc740 \uc790\ub8cc\uc758 \uc8fc\uc694 \ud2b9\uc131\uc778 \uc911\uc2ec\uc758 \uc704\uce58\uc640 \ud37c\uc9d0 \uc815\ub3c4\ub97c \ub4dc\ub7ec\ub0b8\ub2e4. \uc694\uc57d \ud1b5\uacc4\uce58\ub294 \ud06c\uac8c \uc9d1\uc911\uacbd\ud5a5\uce58\uc640 \ubcc0\uc0b0\uc131 \uce21\uc815\uce58\ub85c \uad6c\ubd84\ud560 \uc218 \uc788\ub2e4. \uc9d1\uc911\uacbd\ud5a5\uce58 : \uc790\ub8cc\uc758 \uc911\uc2ec\uc774 \uc5b4\ub514\uc778\uc9c0\ub97c \ub098\ud0c0\ub0b8\ub2e4. \uc911\uc2ec\uc744 \uc5b4\ub5bb\uac8c \uc815\uc758\ud558\ub290\ub0d0\uc5d0 \ub530\ub77c \ud1b5\uacc4\ub7c9\uc774 \ub2ec\ub77c\uc9c4\ub2e4. \ud3c9\uade0(mean), \uc911\uc559\uac12(median), \ucd5c\ube48\uac12(mode), [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":1185,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[32,28,29],"tags":[24,27,25,26],"jetpack_featured_media_url":"http:\/\/ds.sumeun.org\/wp-content\/uploads\/2019\/05\/mosaic_mtcars.png","_links":{"self":[{"href":"http:\/\/ds.sumeun.org\/index.php?rest_route=\/wp\/v2\/posts\/1181"}],"collection":[{"href":"http:\/\/ds.sumeun.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/ds.sumeun.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/ds.sumeun.org\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/ds.sumeun.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1181"}],"version-history":[{"count":5,"href":"http:\/\/ds.sumeun.org\/index.php?rest_route=\/wp\/v2\/posts\/1181\/revisions"}],"predecessor-version":[{"id":1189,"href":"http:\/\/ds.sumeun.org\/index.php?rest_route=\/wp\/v2\/posts\/1181\/revisions\/1189"}],"wp:featuredmedia":[{"embeddable":true,"href":"http:\/\/ds.sumeun.org\/index.php?rest_route=\/wp\/v2\/media\/1185"}],"wp:attachment":[{"href":"http:\/\/ds.sumeun.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1181"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/ds.sumeun.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1181"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/ds.sumeun.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1181"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}