<\/p>\n
Bir veri setinin n’inci<\/sup> y\u00fczdelik<\/strong> dilimi, t\u00fcm de\u011ferler k\u00fc\u00e7\u00fckten b\u00fcy\u00fc\u011fe s\u0131raland\u0131\u011f\u0131nda veri de\u011ferlerinin ilk y\u00fczde n’lik<\/em> k\u0131sm\u0131n\u0131 kesen de\u011ferdir.<\/span><\/p>\n \u00d6rne\u011fin, bir veri k\u00fcmesinin 90’\u0131nc\u0131 y\u00fczdelik dilimi, veri de\u011ferlerinin alt %90’\u0131n\u0131 veri de\u011ferlerinin \u00fcst %10’undan ay\u0131ran de\u011ferdir.<\/span><\/p>\n En s\u0131k kullan\u0131lan y\u00fczdelik dilimlerden biri, bir veri k\u00fcmesinin medyan de\u011ferini temsil eden 50. y\u00fczdelik dilimdir: bu, t\u00fcm veri de\u011ferlerinin %50’sinin alt\u0131na d\u00fc\u015ft\u00fc\u011f\u00fc de\u011ferdir.<\/span><\/p>\n Y\u00fczdelikler a\u015fa\u011f\u0131daki gibi sorular\u0131 yan\u0131tlamak i\u00e7in kullan\u0131labilir:<\/span><\/p>\n A\u015fa\u011f\u0131daki s\u00f6zdizimini kullanan quantile()<\/strong> i\u015flevini kullanarak R’deki y\u00fczdelik dilimleri kolayca hesaplayabiliriz:<\/span><\/p>\n niceliksel<\/strong> (x, olas\u0131l\u0131klar = s\u0131ra(0, 1, 0,25))<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, R’de belirli bir vekt\u00f6r i\u00e7in farkl\u0131 y\u00fczdelik de\u011ferlerin nas\u0131l bulunaca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n Belirli bir veri \u00e7er\u00e7evesi s\u00fctununun y\u00fczdelik dilimlerinin nas\u0131l bulunaca\u011f\u0131n\u0131 g\u00f6stermek i\u00e7in yerle\u015fik veri k\u00fcmesi iris’i<\/em> kullanaca\u011f\u0131z:<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, Sepal.Length<\/em> s\u00fctunu i\u00e7in 90. y\u00fczdelik de\u011ferin nas\u0131l bulunaca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n Apply()<\/strong><\/a> i\u015flevini kullanarak birden fazla s\u00fctunun y\u00fczdelik dilimlerini ayn\u0131 anda da bulabiliriz:<\/span><\/p>\n Ayr\u0131ca dplyr<\/strong><\/a> k\u00fct\u00fcphanesindeki group_by()<\/strong> fonksiyonunu kullanarak R’de gruba g\u00f6re y\u00fczdelik dilimleri de bulabiliriz.<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, her biri i\u00e7in Sepal.Length’in<\/em> 90. y\u00fczdelik diliminin nas\u0131l bulunaca\u011f\u0131n\u0131 g\u00f6sterir.<\/span> A\u015fa\u011f\u0131daki kod, t\u00fcrlere g\u00f6re t\u00fcm de\u011fi\u015fkenler i\u00e7in 90. y\u00fczdelik dilimin nas\u0131l bulunaca\u011f\u0131n\u0131 g\u00f6stermektedir:<\/span><\/p>\n R’de bir veri k\u00fcmesinin y\u00fczdelik dilimlerini g\u00f6rselle\u015ftirmek i\u00e7in yerle\u015fik bir i\u015flev yoktur, ancak y\u00fczdelik dilimlerini nispeten kolay bir \u015fekilde g\u00f6rselle\u015ftirmek i\u00e7in bir grafik olu\u015fturabiliriz.<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, iris<\/em> veri k\u00fcmesinden Sepal.Length<\/em> veri de\u011ferleri i\u00e7in y\u00fczdelik grafi\u011fin nas\u0131l olu\u015fturulaca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n R’de application(), lapply(), sapply() ve tapply() i\u00e7in bir k\u0131lavuz<\/a> Bir veri setinin n’inci y\u00fczdelik dilimi, t\u00fcm de\u011ferler k\u00fc\u00e7\u00fckten b\u00fcy\u00fc\u011fe s\u0131raland\u0131\u011f\u0131nda veri de\u011ferlerinin ilk y\u00fczde n’lik k\u0131sm\u0131n\u0131 kesen de\u011ferdir. \u00d6rne\u011fin, bir veri k\u00fcmesinin 90’\u0131nc\u0131 y\u00fczdelik dilimi, veri de\u011ferlerinin alt %90’\u0131n\u0131 veri de\u011ferlerinin \u00fcst %10’undan ay\u0131ran de\u011ferdir. En s\u0131k kullan\u0131lan y\u00fczdelik dilimlerden biri, bir veri k\u00fcmesinin medyan de\u011ferini temsil eden 50. y\u00fczdelik dilimdir: bu, t\u00fcm veri […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[11],"tags":[],"class_list":["post-513","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n\n
R’de y\u00fczdelikler nas\u0131l hesaplan\u0131r<\/strong><\/span><\/h2>\n
\n
Bir vekt\u00f6r\u00fcn y\u00fczdelik dilimlerini bulma<\/strong><\/span><\/h3>\n
#create vector of 100 random values uniformly distributed between 0 and 500\n<\/span>data <- runif(100, 0, 500)\n\n#Find the quartiles (25th, 50th, and 75th percentiles) of the vector\n<\/span>quantile<\/span> (data, probs = c(.25, .5, .75))\n\n# 25% 50% 75% \n#97.78961 225.07593 356.47943 \n\n#Find the deciles (10th, 20th, 30th, ..., 90th percentiles) of the vector\n<\/span>quantile<\/span> (data, probs = seq(.1, .9, by = .1))\n\n# 10% 20% 30% 40% 50% 60% 70% 80% \n#45.92510 87.16659 129.49574 178.27989 225.07593 300.79690 337.84393 386.36108 \n#90% \n#423.28070\n\n#Find the 37th, 53rd, and 87th percentiles\n<\/span>quantile<\/span> (data, probs = c(.37, .53, .87))\n\n# 37% 53% 87% \n#159.9561 239.8420 418.4787 \n<\/strong><\/pre>\n Veri \u00e7er\u00e7evesi s\u00fctununun y\u00fczdelik dilimlerini bulma<\/strong><\/span><\/h3>\n
#view first six rows of iris<\/em> dataset\n<\/span>head(iris)\n\n Sepal.Length Sepal.Width Petal.Length Petal.Width Species\n1 5.1 3.5 1.4 0.2 setosa\n2 4.9 3.0 1.4 0.2 setosa\n3 4.7 3.2 1.3 0.2 setosa\n4 4.6 3.1 1.5 0.2 setosa\n5 5.0 3.6 1.4 0.2 setosa\n6 5.4 3.9 1.7 0.4 setosa\n<\/strong><\/pre>\n quantile<\/span> (iris$Sepal.Length, probs = 0.9)\n\n#90% \n#6.9<\/strong><\/pre>\n Birden \u00e7ok veri \u00e7er\u00e7evesi s\u00fctununun y\u00fczdelik dilimlerini bulma<\/strong><\/span><\/h3>\n
#define columns we want to find percentiles for\n<\/span>small_iris<- iris[, c('Sepal.Length', 'Sepal.Width', 'Petal.Length', 'Petal.Width')]\n\n#use apply()<\/em> function to find 90th percentile for every column<\/span>\napply<\/span> (small_iris, 2, function(x) quantile(x, probs = .9))\n\n#Sepal.Length Sepal.Width Petal.Length Petal.Width \n#6.90 3.61 5.80 2.20<\/strong><\/pre>\n Y\u00fczdelik dilimleri gruba g\u00f6re arama<\/strong><\/span><\/h3>\n
iris veri setindeki \u00fc\u00e7 t\u00fcr:<\/span><\/p>\n #load dplyr<\/em> library<\/span>\nlibrary(dplyr)\n\n#find 90th percentile of Sepal.Length<\/em> for each of the three species<\/span>\niris %>%\n group_by<\/span> (Species) %>%\n summarize<\/span> (percent90 = quantile(Sepal.Length, probs = .9))\n\n# A tibble: 3 x 2\n#Speciespercent90\n# \n#1 setosa 5.41\n#2 versicolor 6.7 \n#3 virginica 7.61<\/strong><\/pre>\n iris %>%\n group_by<\/span> (Species) %>%\n summarize<\/span> (percent90_SL = quantile(Sepal.Length, probs = .9),\n percent90_SW = quantile(Sepal.Width, probs = .9),\n percent90_PL = quantile(Petal.Length, probs = .9),\n percent90_PW = quantile(Petal.Width, probs = .9))\n\n# A tibble: 3 x 5\n# Species percent90_SL percent90_SW percent90_PL percent90_PW\n# \n#1 setosa 5.41 3.9 1.7 0.4 \n#2 versicolor 6.7 3.11 4.8 1.51\n#3 virginica 7.61 3.31 6.31 2.4 \n<\/strong><\/pre>\n Y\u00fczdelikler g\u00f6r\u00fcnt\u00fcleniyor<\/strong><\/span><\/h2>\n
n = length(iris$Sepal.Length)\nplot((1:n - 1)\/(n - 1), sort(iris$Sepal.Length), type=\"l\",\n main = \"Visualizing Percentiles\",\n xlab = \"Percentile\",\n ylab = \"Value\")<\/strong><\/pre>\n
Ek kaynaklar<\/b><\/span><\/h3>\n
Mutate() ve case_when() ile R’de yeni de\u011fi\u015fkenler olu\u015fturun<\/a>
<\/strong><\/span><\/p>\n","protected":false},"excerpt":{"rendered":"