<\/p>\n
Medyan mutlak sapma,<\/strong> bir veri setindeki g\u00f6zlemlerin<\/a> da\u011f\u0131l\u0131m\u0131n\u0131 \u00f6l\u00e7er.<\/span><\/p>\n Bu \u00f6zellikle yararl\u0131 bir \u00f6l\u00e7\u00fcmd\u00fcr \u00e7\u00fcnk\u00fc standart sapma ve varyans gibi di\u011fer da\u011f\u0131l\u0131m \u00f6l\u00e7\u00fcmlerine g\u00f6re ayk\u0131r\u0131 de\u011ferlerden daha az etkilenir.<\/span><\/p>\n Genellikle MAD olarak k\u0131salt\u0131lan mutlak medyan sapmay\u0131 hesaplama form\u00fcl\u00fc \u015f\u00f6yledir:<\/span><\/p>\n MAD = medyan (|x i<\/sub> \u2013 x m<\/sub> |)<\/strong><\/span><\/p>\n Alt\u0131n:<\/span><\/p>\n A\u015fa\u011f\u0131daki \u00f6rnekler Python’da statsmodels<\/a> mad<\/strong> i\u015flevini kullanarak medyan mutlak sapman\u0131n nas\u0131l hesaplanaca\u011f\u0131n\u0131 g\u00f6sterir.<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, Python’da tek bir NumPy dizisi i\u00e7in medyan mutlak sapman\u0131n nas\u0131l hesaplanaca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n Veri k\u00fcmesinin medyan mutlak sapmas\u0131 11,1195<\/strong> olarak ortaya \u00e7\u0131k\u0131yor.<\/span><\/p>\n MAD’yi hesaplamak i\u00e7in kullan\u0131lan form\u00fcl\u00fcn, sonucu yakla\u015f\u0131k 0,67 fakt\u00f6r\u00fcyle \u00f6l\u00e7eklendirerek normal da\u011f\u0131l\u0131m varsayarak standart sapman\u0131n sa\u011flam bir tahminini hesaplad\u0131\u011f\u0131n\u0131 unutmamak \u00f6nemlidir.<\/span><\/p>\n Bu \u00f6l\u00e7eklendirme fakt\u00f6r\u00fcn\u00fc kullanmaktan ka\u00e7\u0131nmak i\u00e7in c = 1’i a\u015fa\u011f\u0131daki gibi ayarlaman\u0131z yeterlidir:<\/span><\/p>\n A\u015fa\u011f\u0131daki kod, pandas DataFrame’deki tek bir s\u00fctun i\u00e7in MAD’nin nas\u0131l hesaplanaca\u011f\u0131n\u0131 g\u00f6sterir:<\/span><\/p>\n B<\/em> s\u00fctununun mutlak medyan sapmas\u0131 2,965204<\/strong> olarak \u00e7\u0131k\u0131yor.<\/span><\/p>\n Pandalar\u0131n DataFrame’indeki birden fazla s\u00fctun i\u00e7in MAD’yi hesaplamak amac\u0131yla benzer s\u00f6zdizimini kullanabiliriz:<\/span><\/p>\n Mutlak medyan sapma A s\u00fctunu i\u00e7in 0<\/b> , B s\u00fctunu i\u00e7in 2,965204<\/strong> ve C s\u00fctunu i\u00e7in 2,965204’t\u00fcr<\/b> .<\/span><\/p>\n Python’da MAPE nas\u0131l hesaplan\u0131r<\/a> Medyan mutlak sapma, bir veri setindeki g\u00f6zlemlerin da\u011f\u0131l\u0131m\u0131n\u0131 \u00f6l\u00e7er. Bu \u00f6zellikle yararl\u0131 bir \u00f6l\u00e7\u00fcmd\u00fcr \u00e7\u00fcnk\u00fc standart sapma ve varyans gibi di\u011fer da\u011f\u0131l\u0131m \u00f6l\u00e7\u00fcmlerine g\u00f6re ayk\u0131r\u0131 de\u011ferlerden daha az etkilenir. Genellikle MAD olarak k\u0131salt\u0131lan mutlak medyan sapmay\u0131 hesaplama form\u00fcl\u00fc \u015f\u00f6yledir: MAD = medyan (|x i \u2013 x m |) Alt\u0131n: x i : veri k\u00fcmesinin i’inci […]<\/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-1283","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n\n
\u00d6rnek 1: Bir tablo i\u00e7in MAD’yi hesaplay\u0131n<\/strong><\/span><\/h3>\n
import<\/span> numpy as<\/span> np\nfrom<\/span> statsmodels import<\/span> robust<\/span>\n\n#define data\n<\/span>data = np. array<\/span> ([1, 4, 4, 7, 12, 13, 16, 19, 22, 24])\n\n#calculate MAD\n<\/span>robust. mad<\/span> (data)\n\n11.1195\n<\/strong><\/pre>\n #calculate MAD without scaling factor\n<\/span>robust. mad<\/span> (data, c=1)\n\n7.5<\/strong><\/pre>\n \u00d6rnek 2: DataFrame i\u00e7in MAD’yi hesaplama<\/strong><\/span><\/h3>\n
#make this example reproducible\n<\/span>n.p. random<\/span> . seeds<\/span> (1)\n\n#create pandas DataFrame\n<\/span>data = pd. DataFrame<\/span> ( np.random.randint<\/span> (0,10,size=(5,3)),columns =<\/span> [' A<\/span> ',' B<\/span> ',' C<\/span> '])\n\n#view DataFrame\n<\/span>data\n\n A B C\n0 5 8 9\n1 5 0 0\n2 1 7 6\n3 9 2 4\n4 5 2 4\n\n#calculate MAD for column B<\/em>\ndata[[' B<\/span> ']]. apply<\/span> (robust. mad<\/span> )\n\nB 2.965204\ndtype:float64<\/span><\/span><\/strong><\/pre>\n #calculate MAD for all columns<\/span>\ndata[[' A<\/span> ',' B<\/span> ',' C<\/span> ']]. apply<\/span> (robust. mad<\/span> )\n\nAt 0.000000\nB 2.965204\nC 2.965204\ndtype:float64\n<\/span><\/span><\/strong><\/pre>\n Ek kaynaklar<\/strong><\/span><\/h3>\n
Python’da SMAPE nas\u0131l hesaplan\u0131r<\/a>
Python’da RMSE Nas\u0131l Hesaplan\u0131r?<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"