<\/p>\n
Do\u011frusal regresyonun temel varsay\u0131mlar\u0131ndan<\/a> biri, art\u0131klar\u0131n<\/a> , yorday\u0131c\u0131 de\u011fi\u015fkenin her d\u00fczeyinde e\u015fit varyansla da\u011f\u0131ld\u0131\u011f\u0131d\u0131r. Bu varsay\u0131ma e\u015f varyansl\u0131l\u0131k<\/strong> denir.<\/span><\/p>\n Bu varsay\u0131ma uyulmad\u0131\u011f\u0131 takdirde art\u0131klarda de\u011fi\u015fen varyans\u0131n<\/a> mevcut oldu\u011fu s\u00f6ylenir. Bu durumda regresyon sonu\u00e7lar\u0131 g\u00fcvenilmez hale gelir.<\/span><\/p>\n Bu sorunu \u00e7\u00f6zmenin bir yolu, a\u011f\u0131rl\u0131kl\u0131 en k\u00fc\u00e7\u00fck kareler regresyonunu<\/strong> kullanmakt\u0131r; bu regresyon, g\u00f6zlemlere<\/a> a\u011f\u0131rl\u0131klar atayarak, d\u00fc\u015f\u00fck hata varyans\u0131na sahip olanlar\u0131n, daha b\u00fcy\u00fck hata varyans\u0131na sahip g\u00f6zlemlere k\u0131yasla daha fazla bilgi i\u00e7erdiklerinden daha fazla a\u011f\u0131rl\u0131k almas\u0131n\u0131 sa\u011flar.<\/span><\/p>\n Bu e\u011fitimde Python’da a\u011f\u0131rl\u0131kl\u0131 en k\u00fc\u00e7\u00fck kareler regresyonunun nas\u0131l ger\u00e7ekle\u015ftirilece\u011fine ili\u015fkin ad\u0131m ad\u0131m bir \u00f6rnek sunulmaktad\u0131r.<\/span><\/p>\n \u00d6ncelikle bir s\u0131n\u0131ftaki 16 \u00f6\u011frencinin ders saati say\u0131s\u0131 ve final s\u0131nav notu bilgilerini i\u00e7eren a\u015fa\u011f\u0131daki panda DataFrame’i olu\u015ftural\u0131m:<\/span><\/p>\n Daha sonra, tahmin de\u011fi\u015fkeni olarak saatleri<\/strong> ve yan\u0131t de\u011fi\u015fkeni olarak puan\u0131<\/strong> kullanarak basit bir do\u011frusal regresyon modeline uyum sa\u011flamak i\u00e7in statsmodels<\/strong> mod\u00fcl\u00fcndeki i\u015flevleri kullanaca\u011f\u0131z:<\/span><\/p>\n Model \u00f6zetinden modelin R-kare de\u011ferinin 0,630<\/strong> oldu\u011funu g\u00f6r\u00fcyoruz.<\/span><\/p>\n \u0130lgili:<\/strong> \u0130yi bir R-kare de\u011feri nedir?<\/a><\/span><\/p>\n Daha sonra, a\u011f\u0131rl\u0131klar\u0131 daha d\u00fc\u015f\u00fck varyansa sahip g\u00f6zlemlerin daha fazla a\u011f\u0131rl\u0131k alaca\u011f\u0131 \u015fekilde ayarlayarak a\u011f\u0131rl\u0131kl\u0131 en k\u00fc\u00e7\u00fck kareler ger\u00e7ekle\u015ftirmek i\u00e7in statsmodels<\/strong> WLS()<\/strong> i\u015flevini kullanabiliriz:<\/span><\/p>\n Sonu\u00e7tan bu a\u011f\u0131rl\u0131kl\u0131 en k\u00fc\u00e7\u00fck kareler modeli i\u00e7in R-kare de\u011ferinin 0,676’ya<\/strong> y\u00fckseldi\u011fini g\u00f6rebiliriz.<\/span><\/p>\n Bu, a\u011f\u0131rl\u0131kl\u0131 en k\u00fc\u00e7\u00fck kareler modelinin, basit do\u011frusal regresyon modeline g\u00f6re s\u0131nav puanlar\u0131ndaki varyans\u0131 daha fazla a\u00e7\u0131klayabildi\u011fini g\u00f6sterir.<\/span><\/p>\n Bu bize a\u011f\u0131rl\u0131kl\u0131 en k\u00fc\u00e7\u00fck kareler modelinin basit do\u011frusal regresyon modeline k\u0131yasla verilere daha iyi uyum sa\u011flad\u0131\u011f\u0131n\u0131 s\u00f6yl\u00fcyor.<\/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 Art\u0131k Arsa Nas\u0131l Olu\u015fturulur<\/a> Do\u011frusal regresyonun temel varsay\u0131mlar\u0131ndan biri, art\u0131klar\u0131n , yorday\u0131c\u0131 de\u011fi\u015fkenin her d\u00fczeyinde e\u015fit varyansla da\u011f\u0131ld\u0131\u011f\u0131d\u0131r. Bu varsay\u0131ma e\u015f varyansl\u0131l\u0131k denir. Bu varsay\u0131ma uyulmad\u0131\u011f\u0131 takdirde art\u0131klarda de\u011fi\u015fen varyans\u0131n mevcut oldu\u011fu s\u00f6ylenir. Bu durumda regresyon sonu\u00e7lar\u0131 g\u00fcvenilmez hale gelir. Bu sorunu \u00e7\u00f6zmenin bir yolu, a\u011f\u0131rl\u0131kl\u0131 en k\u00fc\u00e7\u00fck kareler regresyonunu kullanmakt\u0131r; bu regresyon, g\u00f6zlemlere a\u011f\u0131rl\u0131klar atayarak, d\u00fc\u015f\u00fck hata varyans\u0131na […]<\/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-3818","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n 1. Ad\u0131m: Verileri olu\u015fturun<\/strong><\/span><\/h2>\n
import<\/span> pandas as<\/span> pd\n\n#createDataFrame\n<\/span>df = pd. DataFrame<\/span> ({' hours<\/span> ': [1, 1, 2, 2, 2, 3, 4, 4, 4, 5, 5, 5, 6, 6, 7, 8],\n ' score<\/span> ': [48, 78, 72, 70, 66, 92, 93, 75, 75, 80, 95, 97,\n 90, 96, 99, 99]})\n\n#view first five rows of DataFrame\n<\/span>print<\/span> ( df.head<\/span> ())\n\n hours score\n0 1 48\n1 1 78\n2 2 72\n3 2 70\n4 2 66<\/strong><\/pre>\n Ad\u0131m 2: Basit do\u011frusal regresyon modelini yerle\u015ftirin<\/strong><\/span><\/h2>\n
import<\/span> statsmodels.api as<\/span> sm\n\n#define predictor and response variables\n<\/span>y = df[' score<\/span> ']\nX = df[' hours<\/span> ']\n\n#add constant to predictor variables\n<\/span>X = sm. add_constant<\/span> (x)\n\n#fit linear regression model\n<\/span>fit = sm. OLS<\/span> (y,x). fit<\/span> ()\n\n#view model summary\n<\/span>print<\/span> ( fit.summary<\/span> ())\n\n OLS Regression Results \n==================================================== ============================\nDept. Variable: R-squared score: 0.630\nModel: OLS Adj. R-squared: 0.603\nMethod: Least Squares F-statistic: 23.80\nDate: Mon, 31 Oct 2022 Prob (F-statistic): 0.000244\nTime: 11:19:54 Log-Likelihood: -57.184\nNo. Observations: 16 AIC: 118.4\nDf Residuals: 14 BIC: 119.9\nModel: 1 \nCovariance Type: non-robust \n==================================================== ============================\n coef std err t P>|t| [0.025 0.975]\n-------------------------------------------------- ----------------------------\nconst 60.4669 5.128 11.791 0.000 49.468 71.465\nhours 5.5005 1.127 4.879 0.000 3.082 7.919\n==================================================== ============================\nOmnibus: 0.041 Durbin-Watson: 1.910\nProb(Omnibus): 0.980 Jarque-Bera (JB): 0.268\nSkew: -0.010 Prob(JB): 0.875\nKurtosis: 2.366 Cond. No. 10.5<\/strong><\/pre>\n Ad\u0131m 3: A\u011f\u0131rl\u0131kland\u0131r\u0131lm\u0131\u015f en k\u00fc\u00e7\u00fck kareler modelini yerle\u015ftirin<\/strong><\/span><\/h2>\n
#define weights to use\n<\/span>wt = 1\/smf. ols<\/span> (' fit.resid.abs() ~ fit.fittedvalues<\/span> ', data=df). fit<\/span> (). fitted values<\/span> **2\n\n#fit weighted least squares regression model\n<\/span>fit_wls = sm. WLS<\/span> (y, X, weights=wt). fit<\/span> ()\n\n#view summary of weighted least squares regression model\n<\/span>print<\/span> ( fit_wls.summary<\/span> ())\n\n WLS Regression Results \n==================================================== ============================\nDept. Variable: R-squared score: 0.676\nModel: WLS Adj. R-squared: 0.653\nMethod: Least Squares F-statistic: 29.24\nDate: Mon, 31 Oct 2022 Prob (F-statistic): 9.24e-05\nTime: 11:20:10 Log-Likelihood: -55.074\nNo. Comments: 16 AIC: 114.1\nDf Residuals: 14 BIC: 115.7\nModel: 1 \nCovariance Type: non-robust \n==================================================== ============================\n coef std err t P>|t| [0.025 0.975]\n-------------------------------------------------- ----------------------------\nconst 63.9689 5.159 12.400 0.000 52.905 75.033\nhours 4.7091 0.871 5.407 0.000 2.841 6.577\n==================================================== ============================\nOmnibus: 2,482 Durbin-Watson: 1,786\nProb(Omnibus): 0.289 Jarque-Bera (JB): 1.058\nSkew: 0.029 Prob(JB): 0.589\nKurtosis: 1.742 Cond. No. 17.6\n==================================================== ============================<\/strong><\/span><\/pre>\n Ek kaynaklar<\/strong><\/span><\/h2>\n
Python’da QQ Grafi\u011fi Nas\u0131l Olu\u015fturulur<\/a>
Python’da \u00e7oklu do\u011frusall\u0131k nas\u0131l test edilir<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"