Ancak \u00e7oklu do\u011frusal regresyonun temel varsay\u0131mlar\u0131ndan<\/a> biri, her yorday\u0131c\u0131 de\u011fi\u015fken ile yan\u0131t de\u011fi\u015fkeni aras\u0131nda do\u011frusal bir ili\u015fkinin olmas\u0131d\u0131r.<\/span><\/p>\n Bu varsay\u0131m kar\u015f\u0131lanmazsa regresyon modelinin sonu\u00e7lar\u0131 g\u00fcvenilir olmayabilir.<\/span><\/p>\n Bu varsay\u0131m\u0131 test etmenin bir yolu, bir yorday\u0131c\u0131 de\u011fi\u015fkenin yan\u0131t de\u011fi\u015fkenine g\u00f6re art\u0131klar\u0131n\u0131<\/a> g\u00f6r\u00fcnt\u00fcleyen bir k\u0131smi art\u0131klar grafi\u011fi<\/strong> olu\u015fturmakt\u0131r.<\/span><\/p>\n A\u015fa\u011f\u0131daki \u00f6rnek, R’de bir regresyon modeli i\u00e7in k\u0131smi art\u0131k grafiklerinin nas\u0131l olu\u015fturulaca\u011f\u0131n\u0131 g\u00f6sterir.<\/span><\/p>\n R’de \u00fc\u00e7 \u00f6ng\u00f6r\u00fcc\u00fc de\u011fi\u015fkene sahip bir regresyon modeline uydu\u011fumuzu varsayal\u0131m:<\/span><\/p>\n Modeldeki her tahmin de\u011fi\u015fkeni i\u00e7in k\u0131smi art\u0131k grafikleri olu\u015fturmak amac\u0131yla R’deki araba<\/strong> paketindeki crPlots()<\/strong> i\u015flevini kullanabiliriz:<\/span> <\/p>\n Mavi \u00e7izgi, yorday\u0131c\u0131 ile yan\u0131t de\u011fi\u015fkeni aras\u0131ndaki ili\u015fkinin do\u011frusal olmas\u0131 durumunda beklenen art\u0131klar\u0131 g\u00f6sterir. Pembe \u00e7izgi ger\u00e7ek art\u0131klar\u0131 g\u00f6sterir.<\/span><\/p>\n \u0130ki \u00e7izgi \u00f6nemli \u00f6l\u00e7\u00fcde farkl\u0131ysa, bu do\u011frusal olmayan bir ili\u015fkiye i\u015faret eder.<\/span><\/p>\n Yukar\u0131daki grafiklerden x2 ve x3’e ait art\u0131klar\u0131n do\u011frusal olmad\u0131\u011f\u0131n\u0131 g\u00f6rebiliriz.<\/span><\/p>\n Bu, \u00e7oklu do\u011frusal regresyonun do\u011frusall\u0131k varsay\u0131m\u0131n\u0131 ihlal eder. Bu sorunu \u00e7\u00f6zmenin bir yolu, tahmin de\u011fi\u015fkenleri \u00fczerinde kare veya k\u00fcbik k\u00f6k d\u00f6n\u00fc\u015f\u00fcm\u00fc kullanmakt\u0131r:<\/span> <\/p>\n K\u0131smi art\u0131k grafiklerinden x2’nin art\u0131k yan\u0131t de\u011fi\u015fkeniyle daha do\u011frusal bir ili\u015fkiye sahip oldu\u011funu g\u00f6rebiliriz.<\/span><\/p>\n x3 tahmin de\u011fi\u015fkeni hala bir \u015fekilde do\u011frusal de\u011fildir, dolay\u0131s\u0131yla ba\u015fka bir d\u00f6n\u00fc\u015f\u00fcm denemeye veya muhtemelen de\u011fi\u015fkeni modelden tamamen \u00e7\u0131karmaya karar verebiliriz.<\/span><\/p>\n A\u015fa\u011f\u0131daki e\u011fitimlerde R’de di\u011fer ortak grafiklerin nas\u0131l olu\u015fturulaca\u011f\u0131 a\u00e7\u0131klanmaktad\u0131r:<\/span><\/p>\n R’de te\u015fhis grafikleri nas\u0131l olu\u015fturulur<\/a> \u00c7oklu do\u011frusal regresyon, birden fazla yorday\u0131c\u0131 de\u011fi\u015fken ile bir yan\u0131t de\u011fi\u015fkeni aras\u0131ndaki ili\u015fkiyi anlamak i\u00e7in kullanabilece\u011fimiz istatistiksel bir y\u00f6ntemdir. Ancak \u00e7oklu do\u011frusal regresyonun temel varsay\u0131mlar\u0131ndan biri, her yorday\u0131c\u0131 de\u011fi\u015fken ile yan\u0131t de\u011fi\u015fkeni aras\u0131nda do\u011frusal bir ili\u015fkinin olmas\u0131d\u0131r. Bu varsay\u0131m kar\u015f\u0131lanmazsa regresyon modelinin sonu\u00e7lar\u0131 g\u00fcvenilir olmayabilir. Bu varsay\u0131m\u0131 test etmenin bir yolu, bir yorday\u0131c\u0131 de\u011fi\u015fkenin yan\u0131t […]<\/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-2536","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n \u00d6rnek: R’de k\u0131smi art\u0131k grafikleri nas\u0131l olu\u015fturulur?<\/strong><\/span><\/h3>\n
#make this example reproducible\n<\/span>set. seeds<\/span> (0)\n\n#define response variable\n<\/span>y <- c(1:1000)\n\n#define three predictor variables\n<\/span>x1 <- c(1:1000)*runif(n=1000)\nx2 <- (c(1:1000)*rnorm(n=1000))^2\nx3 <- (c(1:1000)*rnorm(n=1000))^3\n\n#fit multiple linear regression model\n<\/span>model <- lm(y~x1+x2+x3))<\/strong><\/pre>\n library<\/span> (car)\n\n#create partial residual plots\n<\/span>crPlots(model)\n<\/strong><\/pre>\n![\"R'deki](\"https:\/\/statorials.org\/wp-content\/uploads\/2023\/08\/partiel1.jpg\")
<\/p>\n library<\/span> (car)\n\n#fit new model with transformed predictor variables\n<\/span>model_transformed <- lm(y~x1+sqrt(x2)+log10(x3^(1\/3)))\n\n#create partial residual plots for new model\n<\/span>crPlots(model_transformed)\n<\/strong><\/pre>\n![\"\"](\"https:\/\/statorials.org\/wp-content\/uploads\/2023\/08\/partiel2.jpg\")
<\/p>\n Ek kaynaklar<\/strong><\/span><\/h3>\n
R’de \u00f6l\u00e7ek ve konum grafi\u011fi nas\u0131l olu\u015fturulur<\/a>
R’de art\u0131k arsa nas\u0131l olu\u015fturulur<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"