{"id":1314,"date":"2023-07-26T22:05:53","date_gmt":"2023-07-26T22:05:53","guid":{"rendered":"https:\/\/statorials.org\/tr\/rde-coklu-dogrusal-regresyonun-grafigini-cizin\/"},"modified":"2023-07-26T22:05:53","modified_gmt":"2023-07-26T22:05:53","slug":"rde-coklu-dogrusal-regresyonun-grafigini-cizin","status":"publish","type":"post","link":"https:\/\/statorials.org\/tr\/rde-coklu-dogrusal-regresyonun-grafigini-cizin\/","title":{"rendered":"R'de \u00e7oklu do\u011frusal regresyon sonu\u00e7lar\u0131 nas\u0131l \u00e7izilir"},"content":{"rendered":"

<\/p>\n

\n

R’de basit do\u011frusal regresyon<\/a> uygulad\u0131\u011f\u0131m\u0131zda, uygun regresyon \u00e7izgisini g\u00f6rselle\u015ftirmek kolayd\u0131r \u00e7\u00fcnk\u00fc yaln\u0131zca tek bir tahmin de\u011fi\u015fkeni ve tek bir yan\u0131t de\u011fi\u015fkeni<\/a> ile \u00e7al\u0131\u015f\u0131yoruz.<\/span><\/p>\n

\u00d6rne\u011fin, a\u015fa\u011f\u0131daki kod basit bir do\u011frusal regresyon modelinin bir veri k\u00fcmesine nas\u0131l s\u0131\u011fd\u0131r\u0131laca\u011f\u0131n\u0131 ve sonu\u00e7lar\u0131n grafi\u011finin nas\u0131l \u00e7izilece\u011fini g\u00f6sterir:<\/span> <\/p>\n

 #create dataset\n<\/span>data <- data.frame(x = c(1, 1, 2, 4, 4, 5, 6, 7, 7, 8, 9, 10, 11, 11),\n                   y = c(13, 14, 17, 23, 24, 25, 25, 24, 28, 32, 33, 35, 40, 41))\n\n#fit simple linear regression model\n<\/span>model <- lm(y ~ x, data = data)\n\n#create scatterplot of data\n<\/span>plot(data$x, data$y)\n\n#add fitted regression line\n<\/span>abline(model)\n<\/strong><\/pre>\n
\"Da\u011f\u0131l\u0131m<\/p>\n
 Bunun yerine, modeldeki di\u011fer \u00f6ng\u00f6r\u00fcc\u00fc de\u011fi\u015fkenlerin varl\u0131\u011f\u0131n\u0131 kontrol ederken<\/em> yan\u0131t de\u011fi\u015fkeni ile bir yorday\u0131c\u0131 de\u011fi\u015fken aras\u0131ndaki ili\u015fkiyi g\u00f6r\u00fcnt\u00fcleyen ayr\u0131 grafikler olan eklenen de\u011fi\u015fken grafiklerini<\/strong> (bazen “k\u0131smi regresyon grafikleri” olarak da adland\u0131r\u0131l\u0131r) kullanabiliriz.<\/span><\/p>\n
 A\u015fa\u011f\u0131daki \u00f6rnek, R’de \u00e7oklu do\u011frusal regresyonun nas\u0131l ger\u00e7ekle\u015ftirilece\u011fini ve eklenen de\u011fi\u015fken grafiklerini kullanarak sonu\u00e7lar\u0131n nas\u0131l g\u00f6rselle\u015ftirilece\u011fini g\u00f6sterir.<\/span><\/p>\n
 \u00d6rnek: R’de \u00c7oklu Do\u011frusal Regresyon Sonu\u00e7lar\u0131n\u0131n Grafi\u011fi<\/strong><\/span><\/h3>\n
 Yerle\u015fik mtcars<\/strong> veri k\u00fcmesini kullanarak a\u015fa\u011f\u0131daki \u00e7oklu do\u011frusal regresyon modelini R’deki bir veri k\u00fcmesine uydurdu\u011fumuzu varsayal\u0131m:<\/span><\/p>\n
 #fit multiple linear regression model\nmodel <- lm(mpg ~ disp + hp + drat, data = mtcars)\n<\/span>\n#view results of model\nsummary(model)\n\nCall:\nlm(formula = mpg ~ disp + hp + drat, data = mtcars)\n\nResiduals:\n    Min 1Q Median 3Q Max \n-5.1225 -1.8454 -0.4456 1.1342 6.4958 \n\nCoefficients:\n             Estimate Std. Error t value Pr(>|t|)   \n(Intercept) 19.344293 6.370882 3.036 0.00513 **\navailable -0.019232 0.009371 -2.052 0.04960 * \nhp -0.031229 0.013345 -2.340 0.02663 * \ndrat 2.714975 1.487366 1.825 0.07863 . \n---\nSignificant. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1\n\nResidual standard error: 3.008 on 28 degrees of freedom\nMultiple R-squared: 0.775, Adjusted R-squared: 0.7509 \nF-statistic: 32.15 on 3 and 28 DF, p-value: 3.28e-09\n<\/span><\/span><\/strong><\/pre>\n
 Sonu\u00e7lardan her bir katsay\u0131 i\u00e7in p de\u011ferinin 0,1’den k\u00fc\u00e7\u00fck oldu\u011funu g\u00f6rebiliriz. Basitlik a\u00e7\u0131s\u0131ndan, yorday\u0131c\u0131 de\u011fi\u015fkenlerin her birinin anlaml\u0131 oldu\u011funu ve modele dahil edilmesi gerekti\u011fini varsayaca\u011f\u0131z.<\/span><\/p>\n
 Eklenen de\u011fi\u015fkenlerin grafiklerini olu\u015fturmak i\u00e7in araba<\/strong> paketindeki avPlots()<\/strong> fonksiyonunu kullanabiliriz:<\/span> <\/p>\n
 #load car package\nlibrary(car)\n<\/span>\n#produce added variable plots\navPlots(model)<\/span>\n<\/span><\/strong><\/pre>\n
\"R'de<\/p>\n
 Her grafi\u011fi nas\u0131l yorumlayaca\u011f\u0131n\u0131z a\u015fa\u011f\u0131da a\u00e7\u0131klanm\u0131\u015ft\u0131r:<\/span><\/p>\n
 Ancak \u00e7oklu do\u011frusal regresyon<\/a> uygulad\u0131\u011f\u0131m\u0131zda sonu\u00e7lar\u0131 g\u00f6rselle\u015ftirmek zorla\u015f\u0131r \u00e7\u00fcnk\u00fc birden fazla tahmin de\u011fi\u015fkeni vard\u0131r ve 2 boyutlu bir grafik \u00fczerinde basit\u00e7e bir regresyon \u00e7izgisi \u00e7izemeyiz.<\/span><\/p>\n