A Comparative Analysis of Ordinary Least Squares and Quantile Regression Estimation Technique
Runyi Emmanuel Francis,
Maureen Tobe Nwakuya
Issue:
Volume 7, Issue 4, December 2022
Pages:
49-54
Received:
30 October 2022
Accepted:
18 November 2022
Published:
30 November 2022
DOI:
10.11648/j.ajmcm.20220704.11
Downloads:
Views:
Abstract: This study explores quantile regression estimation technique and its practicality in regression analysis; hence we provide a comparative study in view of quantile regression as an alternative to the traditional ordinary least squares regression. Although the ordinary least squares (OLS) model examines the relationship between the independent variable and the conditional mean of the dependent variable, whereas the quantile regression model examines the relationship between the independent variable and the conditional quantiles of the dependent variable. Quantile regression overcomes various problems associated with OLS. First, quantile regression is defined and its advantages over ordinary least squares regression are illustrated. Also, specific comparisons are made between ordinary least squares and quantile regression estimation methods. Lastly, both estimation techniques were applied on a real life data and the results obtained from the analysis of two types of datasets in this study suggests that quantile regression provides a richer characterization of the data giving rise to the impact of a covariate on the entire distribution of the response variables as the effect can be very different for different quantiles. Quantile regression therefore gives an efficient and more complete view of the relationship amongst variables, hence, suitable in examining predictors effects at various locations of the outcome distribution.
Abstract: This study explores quantile regression estimation technique and its practicality in regression analysis; hence we provide a comparative study in view of quantile regression as an alternative to the traditional ordinary least squares regression. Although the ordinary least squares (OLS) model examines the relationship between the independent variab...
Show More
Building a Secured Data Warehouse for a University Staff Management System: A Case Study of Gombe State University, Gombe
Muhammed Kabir Ahmed,
Ahmadu Bappah Muhammad,
Abubakar Adamu,
Aishatu Yahaya Umar
Issue:
Volume 7, Issue 4, December 2022
Pages:
55-60
Received:
11 December 2022
Accepted:
4 January 2023
Published:
17 January 2023
Abstract: Universities are among the several organizations with complex activities regarding staff records, payroll, and staff promotion. This work intends to use Data warehouse as a solution to simplify these complex activities. Due to their ability to combine heterogeneous data from several information sources in a single storage location for querying and analysis, data warehousing is gaining importance in terms of strategic decision making throughout time. Data collection is a long-standing practice among organizations. Building a massive data warehouse enables them store all important and relevant information. This information is available, but very few organizations have been able to use it to make informed decisions. Due to the significant role played by the Data Warehouse (DW) and Data Mining in strategic decision making, this work developed a Data Warehouse system that can be used academic staff promotion in a University. The prototype which demonstrated the benefits of Data Warehouse and sensitizes universities in Nigeria to start binding such facilities into their staff management system for the purpose of establishing effective administrative system. The system was implemented under oracle Data Warehouse Builder and is meant to serve as a repository of data for data mining operations. The system offers high degree of accuracy in predicting the case of promotion, staff-students ratio as well as the budget projection.
Abstract: Universities are among the several organizations with complex activities regarding staff records, payroll, and staff promotion. This work intends to use Data warehouse as a solution to simplify these complex activities. Due to their ability to combine heterogeneous data from several information sources in a single storage location for querying and ...
Show More
