{"id":4495,"date":"2023-07-10T16:08:37","date_gmt":"2023-07-10T16:08:37","guid":{"rendered":"https:\/\/statorials.org\/tr\/baslangictaki-lojistik-regresyonu-yorumlama\/"},"modified":"2023-07-10T16:08:37","modified_gmt":"2023-07-10T16:08:37","slug":"baslangictaki-lojistik-regresyonu-yorumlama","status":"publish","type":"post","link":"https:\/\/statorials.org\/tr\/baslangictaki-lojistik-regresyonu-yorumlama\/","title":{"rendered":"Lojistik regresyon kesi\u015fimi nas\u0131l yorumlan\u0131r (\u00f6rnekle)"},"content":{"rendered":"

<\/p>\n

\n

Lojistik regresyon,<\/a> yan\u0131t de\u011fi\u015fkeni<\/a> ikili oldu\u011funda bir regresyon modeline uymak i\u00e7in kullanabilece\u011fimiz bir y\u00f6ntemdir.<\/span><\/p>\n

Lojistik regresyon modelini uygulad\u0131\u011f\u0131m\u0131zda, model \u00e7\u0131kt\u0131s\u0131ndaki orijinal terim, t\u00fcm yorday\u0131c\u0131 de\u011fi\u015fkenler s\u0131f\u0131ra e\u015fit oldu\u011funda ortaya \u00e7\u0131kan yan\u0131t de\u011fi\u015fkeninin log olas\u0131l\u0131\u011f\u0131n\u0131<\/strong> temsil eder.<\/span><\/p>\n

Ancak log olas\u0131l\u0131klar\u0131n\u0131n yorumlanmas\u0131 zor oldu\u011fundan, genellikle kesmeyi olas\u0131l\u0131k a\u00e7\u0131s\u0131ndan \u00e7er\u00e7eveliyoruz.<\/span><\/p>\n

Modeldeki her yorday\u0131c\u0131 de\u011fi\u015fkenin s\u0131f\u0131r oldu\u011fu g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, yan\u0131t de\u011fi\u015fkeninin ortaya \u00e7\u0131kma olas\u0131l\u0131\u011f\u0131n\u0131 anlamak i\u00e7in a\u015fa\u011f\u0131daki form\u00fcl\u00fc kullanabiliriz:<\/span><\/p>\n

 P = e \u03b2 0<\/sub><\/sup> \/ (1 +e \u03b2 0<\/sub><\/sup> )\n<\/strong><\/pre>\n
 A\u015fa\u011f\u0131daki \u00f6rnek, lojistik regresyon kesi\u015fiminin pratikte nas\u0131l yorumlanaca\u011f\u0131n\u0131 g\u00f6stermektedir.
<\/span><\/p>\n
 \u00d6rnek: Lojistik regresyon kesme noktas\u0131 nas\u0131l yorumlan\u0131r?<\/strong><\/span><\/h2>\n
 Bir \u00f6\u011frencinin bir s\u0131n\u0131ftaki final s\u0131nav\u0131n\u0131 ge\u00e7ip ge\u00e7meyece\u011fini tahmin etmek i\u00e7in cinsiyet<\/strong> ve al\u0131nan uygulama s\u0131navlar\u0131n\u0131n say\u0131s\u0131n\u0131<\/strong> kullanarak bir lojistik regresyon modeli uydurmak istedi\u011fimizi varsayal\u0131m.<\/span><\/p>\n
 Modeli istatistiksel yaz\u0131l\u0131m (R, Python<\/a> , Excel<\/a> veya SAS<\/a> gibi) kullanarak uydurdu\u011fumuzu ve a\u015fa\u011f\u0131daki sonucu ald\u0131\u011f\u0131m\u0131z\u0131 varsayal\u0131m:<\/span> <\/p>\n\n\n\n\n\n\n
<\/th>\n Katsay\u0131 tahmini<\/strong><\/span><\/th>\n Standart hata<\/strong><\/span><\/th>\n Z de\u011feri<\/strong><\/span><\/th>\n P de\u011feri<\/strong><\/span><\/th>\n<\/tr>\n
 Tutmak<\/strong><\/span><\/td>\n -1.34<\/span><\/td>\n 0.23<\/span><\/td>\n 5.83<\/span><\/td>\n <0,001<\/span><\/td>\n<\/tr>\n
 Cinsiyet (Erkek = 1)<\/strong><\/span><\/td>\n -0,56<\/span><\/td>\n 0,25<\/span><\/td>\n 2.24<\/span><\/td>\n 0,03<\/span><\/td>\n<\/tr>\n
 Pratik s\u0131navlar<\/strong><\/span><\/td>\n 1.13<\/span><\/td>\n 0,43<\/span><\/td>\n 2.63<\/span><\/td>\n 0,01<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
 Orijinal terimin -1,34<\/strong> de\u011ferine sahip oldu\u011funu g\u00f6rebiliriz.<\/span><\/p>\n
 Bu, cinsiyet<\/strong> s\u0131f\u0131r oldu\u011funda (yani \u00f6\u011frenci k\u0131z oldu\u011funda) ve uygulamal\u0131 s\u0131navlar<\/strong> s\u0131f\u0131r oldu\u011funda (\u00f6\u011frenci final s\u0131nav\u0131na haz\u0131rlan\u0131rken herhangi bir uygulamal\u0131 s\u0131nava girmemi\u015fse), \u00f6\u011frencinin s\u0131nav\u0131 ge\u00e7me logaritmik ihtimalinin -1,34<\/strong> oldu\u011fu anlam\u0131na gelir. .<\/strong> .<\/span><\/p>\n
 G\u00fcnl\u00fck oranlar\u0131n\u0131 anlamak zor oldu\u011fundan, olaylar\u0131 olas\u0131l\u0131k a\u00e7\u0131s\u0131ndan yeniden yazabiliriz:<\/span><\/p>\n
 \u0130lgili:<\/strong> Lojistik Regresyon Katsay\u0131lar\u0131 Nas\u0131l Yorumlan\u0131r?<\/a><\/span><\/p>\n