Algorithmic approaches in model selection of the air passengers flows data

Algorithm is an important element in any problem solving situation.In statistical modelling strategy, the algorithm provides a step by step process in model building, model testing, choosing the ‘best’ model and even forecasting using the chosen model.Tacit knowledge has contributed to the existenc...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلفون الرئيسيون: Ismail, Suzilah, Yusof, Norhayati, Tuan Muda, Tuan Zalizam
التنسيق: Conference or Workshop Item
اللغة:English
منشور في: 2015
الموضوعات:
الوصول للمادة أونلاين:http://repo.uum.edu.my/15488/1/PID218.pdf
http://repo.uum.edu.my/15488/
http://www.icoci.cms.net.my/proceedings/2015/TOC.html
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
الوصف
الملخص:Algorithm is an important element in any problem solving situation.In statistical modelling strategy, the algorithm provides a step by step process in model building, model testing, choosing the ‘best’ model and even forecasting using the chosen model.Tacit knowledge has contributed to the existence of a huge variability in manual modelling process especially between expert and non-expert modellers.Many algorithms (automated model selection) have been developed to bridge the gap either through single or multiple equation modelling.This study aims to evaluate the forecasting performances of several selected algorithms on air passengers flow data based on Root Mean Square Error (RMSE) and Geometric Root Mean Square Error (GRMSE).The findings show that multiple models selection performed well in one and two step-ahead forecast but was outperformed by single model in three step-ahead forecasts.