Title :
Successful Forecasting for Knowledge Discovery by Statistical Methods
Author_Institution :
Coll. of Grad. Studies, Arabian Gulf Univ., Manama, Bahrain
Abstract :
The importance of statistical forecasting is in its ability to discover hidden knowledge from databases. Forecasting is used successfully in many sectors. In this paper, forecasting is implemented in the educational sector, three time series algorithms are used, namely-simple moving average forecasting, weighted average forecasting and linear trend forecasting. Six secondary schools (from 7th to 9th) were selected in this study and data was collected for 9 years. The results indicate that the linear trend forecasting was the most accurate in predicting the school success ratio compared with the simple moving average forecasting and weighted average forecasting, as far as mean absolute error, mean square error and absolute percentage error are concerned.
Keywords :
data mining; educational institutions; forecasting theory; mean square error methods; statistical analysis; time series; absolute percentage error; educational sector; hidden knowledge discovery; linear trend forecasting; mean absolute error; mean square error; school success ratio; secondary schools; simple moving average forecasting; statistical forecasting; time series algorithms; weighted average forecasting; Accuracy; Educational institutions; Forecasting; Mathematical model; Predictive models; Standards; Time series analysis; linear trend forecasting; school sucessful ratio; simple moving average; time series algorithms; weighted average;
Conference_Titel :
Information Technology: New Generations (ITNG), 2012 Ninth International Conference on
Conference_Location :
Las Vegas, NV
Print_ISBN :
978-1-4673-0798-7
DOI :
10.1109/ITNG.2012.160
