<\/p>\n
Python’un statsmodels<\/a> mod\u00fcl\u00fc, \u00e7e\u015fitli istatistiksel modelleri uyarlaman\u0131za olanak tan\u0131yan \u00e7e\u015fitli i\u015flevler ve s\u0131n\u0131flar sunar.<\/span><\/span><\/p>\n A\u015fa\u011f\u0131daki ad\u0131m ad\u0131m \u00f6rnek, statsmodels i\u015flevlerini kullanarak lojistik regresyonun<\/a> nas\u0131l ger\u00e7ekle\u015ftirilece\u011fini g\u00f6sterir.<\/span><\/p>\n \u00d6ncelikle \u00fc\u00e7 de\u011fi\u015fken i\u00e7eren bir pandas DataFrame olu\u015ftural\u0131m:<\/span><\/p>\n Bir \u00f6\u011frencinin belirli bir s\u0131nav\u0131 ge\u00e7ip ge\u00e7meyece\u011fini tahmin etmek i\u00e7in \u00e7al\u0131\u015f\u0131lan saatleri ve \u00e7al\u0131\u015fma y\u00f6ntemini kullanarak bir lojistik regresyon modeli uygulayaca\u011f\u0131z.<\/span><\/p>\n A\u015fa\u011f\u0131daki kod pandalar DataFrame’in nas\u0131l olu\u015fturulaca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n Daha sonra logit()<\/strong> fonksiyonunu kullanarak lojistik regresyon modelini yerle\u015ftirece\u011fiz:<\/span><\/p>\n \u00c7\u0131kt\u0131n\u0131n<\/strong> katsay\u0131 s\u00fctunundaki de\u011ferler bize s\u0131nav\u0131 ge\u00e7me ihtimalinin logdaki ortalama de\u011fi\u015fimini anlat\u0131r.<\/span><\/p>\n \u00d6rne\u011fin:<\/span><\/span><\/p>\n P>|z|<\/strong> i\u00e7indeki de\u011ferler S\u00fctun, her katsay\u0131 i\u00e7in p de\u011ferlerini temsil eder.<\/span><\/p>\n \u00d6rne\u011fin:<\/span><\/p>\n Lojistik regresyon modelinin kalitesini de\u011ferlendirmek i\u00e7in \u00e7\u0131kt\u0131daki iki \u00f6l\u00e7\u00fcme bakabiliriz:<\/span><\/p>\n 1. Takma ad R-kare<\/strong><\/span><\/p>\n Bu de\u011fer, do\u011frusal regresyon modeli i\u00e7in R-kare de\u011ferinin yerine ge\u00e7ebilir.<\/span><\/p>\n Maksimumla\u015ft\u0131r\u0131lm\u0131\u015f log-olabilirlik fonksiyonunun bo\u015f modelden tam modele oran\u0131 olarak hesaplan\u0131r.<\/span><\/p>\n Bu de\u011fer 0 ila 1 aras\u0131nda de\u011fi\u015febilir ve daha y\u00fcksek de\u011ferler daha iyi model uyumunu g\u00f6sterir.<\/span><\/p>\n Bu \u00f6rnekte s\u00f6zde R-kare de\u011feri 0,1894<\/strong> olup olduk\u00e7a d\u00fc\u015f\u00fckt\u00fcr. Bu bize modelin yorday\u0131c\u0131 de\u011fi\u015fkenlerinin yan\u0131t de\u011fi\u015fkeninin de\u011ferini tahmin etme konusunda pek iyi bir i\u015f \u00e7\u0131karmad\u0131\u011f\u0131n\u0131 s\u00f6yler.<\/span><\/p>\n 2. LLR p de\u011feri<\/strong><\/span><\/p>\n Bu de\u011fer, do\u011frusal regresyon modelinin genel F de\u011feri<\/a> i\u00e7in p de\u011ferinin yerine ge\u00e7ebilir.<\/span><\/p>\n Bu de\u011fer belirli bir e\u015fi\u011fin alt\u0131ndaysa (\u00f6rne\u011fin \u03b1 = 0,05), modelin bir b\u00fct\u00fcn olarak “faydal\u0131” oldu\u011fu ve tahmin de\u011fi\u015fkenleri olmayan bir modelle kar\u015f\u0131la\u015ft\u0131r\u0131ld\u0131\u011f\u0131nda yan\u0131t de\u011fi\u015fkeninin de\u011ferlerini daha iyi tahmin edebilece\u011fi sonucuna varabiliriz.<\/span><\/p>\n Bu \u00f6rnekte LLR’nin p de\u011feri 0,07375’tir<\/strong> . Se\u00e7ti\u011fimiz anlaml\u0131l\u0131k d\u00fczeyine ba\u011fl\u0131 olarak (\u00f6rne\u011fin 0,01, 0,05, 0,1), modelin bir b\u00fct\u00fcn olarak yararl\u0131 oldu\u011fu sonucuna varabiliriz ya da \u00e7\u0131kmayabiliriz.<\/span><\/p>\n A\u015fa\u011f\u0131daki e\u011fitimlerde Python’da di\u011fer genel g\u00f6revlerin nas\u0131l ger\u00e7ekle\u015ftirilece\u011fi a\u00e7\u0131klanmaktad\u0131r:<\/span><\/p>\n Python’da do\u011frusal regresyon nas\u0131l ger\u00e7ekle\u015ftirilir<\/a> Python’un statsmodels mod\u00fcl\u00fc, \u00e7e\u015fitli istatistiksel modelleri uyarlaman\u0131za olanak tan\u0131yan \u00e7e\u015fitli i\u015flevler ve s\u0131n\u0131flar sunar. A\u015fa\u011f\u0131daki ad\u0131m ad\u0131m \u00f6rnek, statsmodels i\u015flevlerini kullanarak lojistik regresyonun nas\u0131l ger\u00e7ekle\u015ftirilece\u011fini g\u00f6sterir. 1. Ad\u0131m: Verileri olu\u015fturun \u00d6ncelikle \u00fc\u00e7 de\u011fi\u015fken i\u00e7eren bir pandas DataFrame olu\u015ftural\u0131m: \u00c7al\u0131\u015f\u0131lan saat (de\u011ferin tamam\u0131) \u00c7al\u0131\u015fma y\u00f6ntemi (y\u00f6ntem A veya B) S\u0131nav sonucu (ge\u00e7ti veya kald\u0131) Bir \u00f6\u011frencinin […]<\/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-3526","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n 1. Ad\u0131m: Verileri olu\u015fturun<\/strong><\/span><\/h2>\n
\n
import<\/span> pandas as<\/span> pd\n\n#createDataFrame\n<\/span>df = pd. DataFrame<\/span> ({' result<\/span> ': [0, 1, 0, 0, 0, 0, 0, 1, 1, 0,\n 0, 1, 1, 1, 0, 1, 1, 1, 1, 1],\n ' hours<\/span> ': [1, 2, 2, 2, 3, 2, 5, 4, 3, 6,\n 5, 8, 8, 7, 6, 7, 5, 4, 8, 9],\n ' method<\/span> ': ['A', 'A', 'A', 'B', 'B', 'B', 'B',\n 'B', 'B', 'A', 'B', 'A', 'B', 'B',\n 'A', 'A', 'B', 'A', 'B', 'A']})\n\n#view first five rows of DataFrame\n<\/span>df. head<\/span> ()\n\n\tresult hours method\n0 0 1 A\n1 1 2 A\n2 0 2 A\n3 0 2 B\n4 0 3 B<\/strong><\/pre>\n Ad\u0131m 2: Lojistik regresyon modelini yerle\u015ftirin<\/strong><\/span><\/h2>\n
import<\/span> statsmodels. formula<\/span> . api<\/span> as<\/span> smf\n\n#fit logistic regression model\n<\/span>model = smf. logit<\/span> (' result~hours+method<\/span> ', data=df). fit<\/span> ()\n\n#view model summary\n<\/span>print<\/span> ( model.summary<\/span> ())\n\nOptimization completed successfully.\n Current function value: 0.557786\n Iterations 5\n Logit Regression Results \n==================================================== ============================\nDept. Variable: result No. Observations: 20\nModel: Logit Df Residuals: 17\nMethod: MLE Df Model: 2\nDate: Mon, 22 Aug 2022 Pseudo R-squ.: 0.1894\nTime: 09:53:35 Log-Likelihood: -11.156\nconverged: True LL-Null: -13.763\nCovariance Type: nonrobust LLR p-value: 0.07375\n==================================================== ============================\n coef std err z P>|z| [0.025 0.975]\n-------------------------------------------------- -----------------------------\nIntercept -2.1569 1.416 -1.523 0.128 -4.932 0.618\nmethod[TB] 0.0875 1.051 0.083 0.934 -1.973 2.148\nhours 0.4909 0.245 2.002 0.045 0.010 0.972\n==================================================== ============================\n<\/strong><\/span><\/pre>\n\n
\n
3. Ad\u0131m: Model performans\u0131n\u0131 de\u011ferlendirin<\/strong><\/span><\/h2>\n
Ek kaynaklar<\/strong><\/span><\/h2>\n
Python’da logaritmik regresyon nas\u0131l ger\u00e7ekle\u015ftirilir<\/a>
Python’da niceliksel regresyon nas\u0131l ger\u00e7ekle\u015ftirilir<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"