A\u015fa\u011f\u0131daki \u015fekilde hesaplan\u0131r:<\/span><\/p>\n MAPE<\/strong> = (1\/n) * \u03a3(|ger\u00e7ek \u2013 tahmin| \/ |ger\u00e7ek|) * 100<\/span><\/p>\n Alt\u0131n:<\/span><\/p>\n MAPE, yorumlanmas\u0131 kolay oldu\u011fundan yayg\u0131n olarak kullan\u0131l\u0131r. \u00d6rne\u011fin<\/span> %14’l\u00fck bir MAPE de\u011feri, tahmin edilen de\u011fer ile ger\u00e7ek de\u011fer aras\u0131ndaki ortalama fark\u0131n %14 oldu\u011fu anlam\u0131na gelir.<\/span><\/p>\n A\u015fa\u011f\u0131daki \u00f6rnek, belirli bir model i\u00e7in MAPE de\u011ferinin nas\u0131l hesaplanaca\u011f\u0131n\u0131 ve yorumlanaca\u011f\u0131n\u0131 g\u00f6sterir.<\/span><\/p>\n Bir bakkal zincirinin gelecekteki sat\u0131\u015flar\u0131 tahmin etmek i\u00e7in bir model olu\u015fturdu\u011funu varsayal\u0131m. A\u015fa\u011f\u0131daki grafik, modelin ard\u0131\u015f\u0131k 12 sat\u0131\u015f d\u00f6nemi i\u00e7in fiili sat\u0131\u015flar\u0131n\u0131 ve tahmini sat\u0131\u015flar\u0131n\u0131 g\u00f6stermektedir:<\/span> <\/p>\n Her tahminin mutlak y\u00fczde hatas\u0131n\u0131 hesaplamak i\u00e7in a\u015fa\u011f\u0131daki form\u00fcl\u00fc kullanabiliriz:<\/span><\/p>\n Daha sonra mutlak hata y\u00fczdelerinin ortalamas\u0131n\u0131 hesaplayabiliriz:<\/span> <\/p>\n Bu modelin MAPE’si %5,12<\/strong> olarak \u00e7\u0131k\u0131yor.<\/span><\/p>\n Bu bize model taraf\u0131ndan tahmin edilen sat\u0131\u015flar ile fiili sat\u0131\u015flar aras\u0131ndaki ortalama mutlak y\u00fczde hatan\u0131n %5,12<\/strong> oldu\u011funu g\u00f6sterir.<\/span><\/p>\n Bunun MAPE i\u00e7in iyi bir de\u011fer<\/a> olup olmad\u0131\u011f\u0131n\u0131n belirlenmesi end\u00fcstri standartlar\u0131na ba\u011fl\u0131d\u0131r.<\/span><\/p>\n Standart g\u0131da end\u00fcstrisi %2’lik bir MAPE de\u011ferine sahip model \u00fcretiyorsa bu %5,12’lik de\u011ferin y\u00fcksek oldu\u011fu d\u00fc\u015f\u00fcn\u00fclebilir.<\/span><\/p>\n Tersine, e\u011fer sekt\u00f6rdeki market tahmin modellerinin \u00e7o\u011fu %10 ile %15 aras\u0131nda MAPE de\u011ferleri \u00fcretiyorsa, o zaman %5,12’lik bir MAPE de\u011feri d\u00fc\u015f\u00fck say\u0131labilir ve bu model gelecekteki sat\u0131\u015flar\u0131 tahmin etmek i\u00e7in m\u00fckemmel olarak de\u011ferlendirilebilir.<\/span><\/p>\n MAPE \u00f6zellikle farkl\u0131 modellerin uyumunu kar\u015f\u0131la\u015ft\u0131rmak i\u00e7in kullan\u0131\u015fl\u0131d\u0131r.<\/span><\/p>\n \u00d6rne\u011fin, bir bakkal zincirinin gelecekteki sat\u0131\u015flar\u0131 tahmin etmek i\u00e7in bir model olu\u015fturmak istedi\u011fini ve \u00e7e\u015fitli potansiyel modeller aras\u0131ndan m\u00fcmk\u00fcn olan en iyi modeli bulmak istedi\u011fini varsayal\u0131m.<\/span><\/p>\n \u00dc\u00e7 farkl\u0131 modeli e\u015fle\u015ftirdiklerini ve kar\u015f\u0131l\u0131k gelen MAPE de\u011ferlerini bulduklar\u0131n\u0131 varsayal\u0131m:<\/span><\/p>\n Model 3 en d\u00fc\u015f\u00fck MAPE de\u011ferine sahiptir ve bu da bize \u00fc\u00e7 potansiyel model aras\u0131nda gelecekteki sat\u0131\u015flar\u0131 en y\u00fcksek do\u011frulukla tahmin edebildi\u011fini s\u00f6yler.<\/span><\/p>\n Excel’de MAPE Nas\u0131l Hesaplan\u0131r?<\/a> Bir modelin tahmin do\u011frulu\u011funu \u00f6l\u00e7mek i\u00e7in en yayg\u0131n kullan\u0131lan \u00f6l\u00e7\u00fcmlerden biri, genellikle MAPE olarak k\u0131salt\u0131lan ortalama mutlak y\u00fczde hatas\u0131d\u0131r . A\u015fa\u011f\u0131daki \u015fekilde hesaplan\u0131r: MAPE = (1\/n) * \u03a3(|ger\u00e7ek \u2013 tahmin| \/ |ger\u00e7ek|) * 100 Alt\u0131n: \u03a3 \u2013 \u201ctoplam\u201d anlam\u0131na gelen bir sembol n \u2013 \u00d6rneklem b\u00fcy\u00fckl\u00fc\u011f\u00fc ger\u00e7ek \u2013 Verinin ger\u00e7ek de\u011feri tahmin \u2013 Verilerin tahmin […]<\/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-1720","post","type-post","status-publish","format-standard","hentry","category-rehber"],"yoast_head":"\n\n
\u00d6rnek: belirli bir model i\u00e7in MAPE de\u011ferini yorumlama<\/strong><\/span><\/h3>\n
![\"\"](\"https:\/\/statorials.org\/wp-content\/uploads\/2023\/08\/mape_interpret1.png\")
<\/p>\n\n
![\"\"](\"https:\/\/statorials.org\/wp-content\/uploads\/2023\/08\/mape_interpret2.png\")
<\/p>\n![\"\"](\"https:\/\/statorials.org\/wp-content\/uploads\/2023\/08\/mape_interpret3.png\")
<\/p>\n Farkl\u0131 modellerin MAPE de\u011ferlerinin kar\u015f\u0131la\u015ft\u0131r\u0131lmas\u0131<\/strong><\/span><\/h3>\n
\n
Ek kaynaklar<\/strong><\/span><\/h3>\n
R’de MAPE nas\u0131l hesaplan\u0131r<\/a>
Python’da MAPE nas\u0131l hesaplan\u0131r<\/a>
MAPE hesaplay\u0131c\u0131s\u0131<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"