<\/p>\n
R’ye do\u011frusal bir regresyon modeli<\/a> s\u0131\u011fd\u0131rmak i\u00e7in lm()<\/strong> komutunu kullanabiliriz.<\/span><\/p>\n Regresyon modelinin \u00e7\u0131kt\u0131s\u0131n\u0131 g\u00f6r\u00fcnt\u00fclemek i\u00e7in Summary()<\/strong> komutunu kullanabiliriz.<\/span><\/p>\n Bu e\u011fitimde, R’deki regresyon \u00e7\u0131kt\u0131s\u0131n\u0131n her de\u011ferinin nas\u0131l yorumlanaca\u011f\u0131 a\u00e7\u0131klanmaktad\u0131r.<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, \u00f6ng\u00f6r\u00fc de\u011fi\u015fkenleri olarak hp<\/em> , drat<\/em> ve wt<\/em> ve yan\u0131t de\u011fi\u015fkeni olarak mpg<\/em> kullan\u0131larak entegre mtcars<\/strong> veri k\u00fcmesiyle \u00e7oklu do\u011frusal regresyon modelinin nas\u0131l s\u0131\u011fd\u0131r\u0131laca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n \u00c7\u0131kt\u0131daki her de\u011ferin nas\u0131l yorumlanaca\u011f\u0131 a\u015fa\u011f\u0131da a\u00e7\u0131klanm\u0131\u015ft\u0131r:<\/span><\/p>\n Bu b\u00f6l\u00fcm bize regresyon modelimizde kulland\u0131\u011f\u0131m\u0131z form\u00fcl\u00fc hat\u0131rlat\u0131yor. Yan\u0131t de\u011fi\u015fkeni olarak mpg’yi<\/strong> , tahmin de\u011fi\u015fkenleri olarak hp<\/strong> , drat<\/strong> ve wt’yi<\/strong> kulland\u0131\u011f\u0131m\u0131z\u0131 g\u00f6rebiliriz. Her de\u011fi\u015fken mtcars<\/strong> ad\u0131 verilen veri k\u00fcmesinden geldi.<\/span><\/p>\n Bu b\u00f6l\u00fcmde regresyon modelinden kalanlar\u0131n da\u011f\u0131l\u0131m\u0131n\u0131n bir \u00f6zeti g\u00f6r\u00fcnt\u00fclenir. Kal\u0131nt\u0131n\u0131n, g\u00f6zlemlenen de\u011fer ile regresyon modelinin tahmin edilen de\u011feri aras\u0131ndaki fark oldu\u011funu hat\u0131rlay\u0131n.<\/span><\/p>\n Minimum kal\u0131nt\u0131 -3,3598<\/strong> , medyan kal\u0131nt\u0131 -0,5099<\/strong> ve maksimum kal\u0131nt\u0131 5,7078<\/strong> idi.<\/span><\/p>\n Bu b\u00f6l\u00fcm regresyon modelinin tahmin edilen katsay\u0131lar\u0131n\u0131 g\u00f6sterir. Bu katsay\u0131lar\u0131 a\u015fa\u011f\u0131daki tahmini regresyon denklemini olu\u015fturmak i\u00e7in kullanabiliriz:<\/span><\/p>\n mpg = 29,39 \u2013 0,03*hp + 1,62*drat \u2013 3,23*a\u011f\u0131rl\u0131k<\/span><\/p>\n Her yorday\u0131c\u0131 de\u011fi\u015fken i\u00e7in a\u015fa\u011f\u0131daki de\u011ferleri al\u0131r\u0131z:<\/span><\/p>\n Tahmin:<\/strong> Tahmin edilen katsay\u0131. Bu bize, di\u011fer t\u00fcm \u00f6ng\u00f6r\u00fcc\u00fc de\u011fi\u015fkenlerin sabit kald\u0131\u011f\u0131n\u0131 varsayarak, yorday\u0131c\u0131 de\u011fi\u015fkendeki bir birimlik art\u0131\u015fla ili\u015fkili yan\u0131t de\u011fi\u015fkenindeki ortalama art\u0131\u015f\u0131 s\u00f6yler.<\/span><\/p>\n Standart.<\/strong> Hata<\/strong> : Bu katsay\u0131n\u0131n standart hatas\u0131d\u0131r. Bu, katsay\u0131ya ili\u015fkin tahminimizin belirsizli\u011finin bir \u00f6l\u00e7\u00fcs\u00fcd\u00fcr.<\/span><\/p>\n t-de\u011feri:<\/strong> (Tahmin) \/ (Standart Hata) olarak hesaplanan, yorday\u0131c\u0131 de\u011fi\u015fkene ait t-istatisti\u011fidir.<\/span><\/p>\n Pr(>|t|):<\/strong> t istatisti\u011fine kar\u015f\u0131l\u0131k gelen p de\u011feridir. Bu de\u011fer belirli bir alfa d\u00fczeyinin alt\u0131ndaysa (\u00f6rne\u011fin 0,05), yorday\u0131c\u0131 de\u011fi\u015fkenin istatistiksel olarak anlaml\u0131 oldu\u011fu s\u00f6ylenir.<\/span><\/p>\n Bu regresyon modelinde hangi yorday\u0131c\u0131lar\u0131n anlaml\u0131 oldu\u011funu belirlemek i\u00e7in \u03b1 = 0,05’lik bir alfa d\u00fczeyi kullan\u0131rsak, hp<\/strong> ve wt’nin<\/strong> istatistiksel olarak anlaml\u0131 yorday\u0131c\u0131lar oldu\u011funu ancak drat’\u0131n<\/strong> anlaml\u0131 olmad\u0131\u011f\u0131n\u0131 s\u00f6ylerdik.<\/span><\/p>\n Bu son b\u00f6l\u00fcm, regresyon modelinin veri setimize ne kadar iyi uydu\u011funu de\u011ferlendirmemize yard\u0131mc\u0131 olan \u00e7e\u015fitli say\u0131lar\u0131 g\u00f6r\u00fcnt\u00fcler.<\/span><\/p>\n Art\u0131k standart hata:<\/strong> Bu bize g\u00f6zlemlenen de\u011ferler ile regresyon \u00e7izgisi aras\u0131ndaki ortalama mesafeyi s\u00f6yler. De\u011fer ne kadar k\u00fc\u00e7\u00fck olursa, regresyon modeli verilere o kadar iyi uyum sa\u011flayabilir.<\/span><\/p>\n Serbestlik dereceleri nk-1 olarak hesaplan\u0131r; burada n = toplam g\u00f6zlem say\u0131s\u0131 ve k = \u00f6ng\u00f6r\u00fcc\u00fclerin say\u0131s\u0131. Bu \u00f6rnekte mtcars’\u0131n 32 g\u00f6zlemi var ve regresyon modelinde 3 \u00f6ng\u00f6r\u00fcc\u00fc kulland\u0131k, dolay\u0131s\u0131yla serbestlik derecesi 32 \u2013 3 \u2013 1 = 28.<\/span><\/p>\n \u00c7oklu R-kare:<\/strong> Buna belirleme katsay\u0131s\u0131 denir. Bize yan\u0131t de\u011fi\u015fkenindeki<\/a> varyans\u0131n ne kadar\u0131n\u0131n yorday\u0131c\u0131 de\u011fi\u015fkenler taraf\u0131ndan a\u00e7\u0131klanabilece\u011fini s\u00f6yler.<\/span><\/p>\n Bu de\u011fer 0 ile 1 aras\u0131nda de\u011fi\u015fir. 1’e ne kadar yak\u0131nsa, yorday\u0131c\u0131 de\u011fi\u015fkenler yan\u0131t de\u011fi\u015fkeninin de\u011ferini o kadar fazla tahmin edebilir.<\/span><\/p>\n D\u00fczeltilmi\u015f R-kare:<\/strong> Bu, modeldeki \u00f6ng\u00f6r\u00fcc\u00fclerin say\u0131s\u0131na g\u00f6re ayarlanan R-karenin de\u011fi\u015ftirilmi\u015f bir versiyonudur. Her zaman R kareden k\u00fc\u00e7\u00fckt\u00fcr.<\/span><\/p>\n D\u00fczeltilmi\u015f R-kare, farkl\u0131 say\u0131da \u00f6ng\u00f6r\u00fcc\u00fc de\u011fi\u015fken kullanan farkl\u0131 regresyon modellerinin uyumunu kar\u015f\u0131la\u015ft\u0131rmak i\u00e7in yararl\u0131 olabilir.<\/span><\/p>\n F-istatisti\u011fi:<\/strong> Regresyon modelinin, ba\u011f\u0131ms\u0131z de\u011fi\u015fken i\u00e7ermeyen bir modele g\u00f6re verilere daha iyi uyum sa\u011flay\u0131p sa\u011flamad\u0131\u011f\u0131n\u0131 g\u00f6sterir. Temel olarak regresyon modelinin bir b\u00fct\u00fcn olarak yararl\u0131 olup olmad\u0131\u011f\u0131n\u0131 test eder.<\/span><\/p>\n p de\u011feri:<\/strong> F istatisti\u011fine kar\u015f\u0131l\u0131k gelen p de\u011feridir. Bu de\u011fer belirli bir anlaml\u0131l\u0131k d\u00fczeyinin alt\u0131ndaysa (\u00f6rne\u011fin 0,05), bu durumda regresyon modeli, yorday\u0131c\u0131 olmayan bir modele g\u00f6re verilere daha iyi uyum sa\u011flar.<\/span><\/p>\n Regresyon modelleri olu\u015ftururken bu p de\u011ferinin belirli bir anlaml\u0131l\u0131k d\u00fczeyinin alt\u0131nda olmas\u0131n\u0131 umuyoruz \u00e7\u00fcnk\u00fc bu, yorday\u0131c\u0131 de\u011fi\u015fkenlerin yan\u0131t de\u011fi\u015fkeninin de\u011ferini tahmin etmede ger\u00e7ekten yararl\u0131 oldu\u011funu g\u00f6sterir.<\/span><\/p>\n R’de basit do\u011frusal regresyon nas\u0131l ger\u00e7ekle\u015ftirilir<\/a> R’ye do\u011frusal bir regresyon modeli s\u0131\u011fd\u0131rmak i\u00e7in lm() komutunu kullanabiliriz. Regresyon modelinin \u00e7\u0131kt\u0131s\u0131n\u0131 g\u00f6r\u00fcnt\u00fclemek i\u00e7in Summary() komutunu kullanabiliriz. Bu e\u011fitimde, R’deki regresyon \u00e7\u0131kt\u0131s\u0131n\u0131n her de\u011ferinin nas\u0131l yorumlanaca\u011f\u0131 a\u00e7\u0131klanmaktad\u0131r. \u00d6rnek: R’de Regresyon \u00c7\u0131kt\u0131s\u0131n\u0131 Yorumlama A\u015fa\u011f\u0131daki kod, \u00f6ng\u00f6r\u00fc de\u011fi\u015fkenleri olarak hp , drat ve wt ve yan\u0131t de\u011fi\u015fkeni olarak mpg kullan\u0131larak entegre mtcars veri k\u00fcmesiyle \u00e7oklu do\u011frusal […]<\/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-1253","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n \u00d6rnek: R’de Regresyon \u00c7\u0131kt\u0131s\u0131n\u0131 Yorumlama<\/strong><\/h3>\n
#fit regression model using hp, drat, and wt as predictors\n<\/span>model <- lm(mpg ~ hp + drat + wt, data = mtcars)\n\n#view model summary\n<\/span>summary(model)\n\nCall:\nlm(formula = mpg ~ hp + drat + wt, data = mtcars)\n\nResiduals:\n Min 1Q Median 3Q Max \n-3.3598 -1.8374 -0.5099 0.9681 5.7078 \n\nCoefficients:\n Estimate Std. Error t value Pr(>|t|) \n(Intercept) 29.394934 6.156303 4.775 5.13e-05 ***\nhp -0.032230 0.008925 -3.611 0.001178 ** \ndrat 1.615049 1.226983 1.316 0.198755 \nwt -3.227954 0.796398 -4.053 0.000364 ***\n---\nSignificant. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1\n\nResidual standard error: 2.561 on 28 degrees of freedom\nMultiple R-squared: 0.8369, Adjusted R-squared: 0.8194 \nF-statistic: 47.88 on 3 and 28 DF, p-value: 3.768e-11\n<\/strong><\/pre>\n Arama<\/strong><\/span><\/h3>\n
Call:\nlm(formula = mpg ~ hp + drat + wt, data = mtcars)<\/strong><\/pre>\n
Kal\u0131nt\u0131<\/strong><\/span><\/h3>\n
Residuals:\n Min 1Q Median 3Q Max \n-3.3598 -1.8374 -0.5099 0.9681 5.7078 \n<\/strong><\/pre>\n
Katsay\u0131lar<\/strong><\/span><\/h3>\n
Coefficients:\n Estimate Std. Error t value Pr(>|t|) \n(Intercept) 29.394934 6.156303 4.775 5.13e-05 ***\nhp -0.032230 0.008925 -3.611 0.001178 ** \ndrat 1.615049 1.226983 1.316 0.198755 \nwt -3.227954 0.796398 -4.053 0.000364 ***\n\n---\nSignificant. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1\n<\/strong><\/pre>\n
Model yeterlili\u011finin de\u011ferlendirilmesi<\/strong><\/span><\/h3>\n
Residual standard error: 2.561 on 28 degrees of freedom\nMultiple R-squared: 0.8369, Adjusted R-squared: 0.8194 \nF-statistic: 47.88 on 3 and 28 DF, p-value: 3.768e-11\n<\/strong><\/pre>\n
Ek kaynaklar<\/strong><\/span><\/h3>\n
R’de \u00e7oklu do\u011frusal regresyon nas\u0131l ger\u00e7ekle\u015ftirilir<\/a>
\u0130yi bir R-kare de\u011feri nedir?<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"