Academic Journal of Applied Mathematical Sciences

Online ISSN: 2415-2188
Print ISSN: 2415-5225

Quarterly Published (4 Issues Per Year)





Archives

Volume 4 Number 9 September 2018

Alphabetic Optimality Criteria for 2K Central Composite Design


Authors: Francis C. Eze ; Lilian Ngonadi O.
Pages: 107-118
Abstract
Designing an appropriate central composite design involves selection of the right type of central composite design. The various Central composite designs(CCDs); Spherical central composite design (SCCD), Rotatable central composite design (RCCD), Orthogonal central composite design (OCCD) and Face centered cube design (FCCD) were compared for factors k =2 to 6 using the D, G and A criteria. The α (axial) values were obtained for the various Central composite design (SCCD, RCCD, OCCD and FCD) and the values used to find relevant results for factors k=2 to 6.  The efficiency values were obtained for the D, A and G optimality criteria in which larger values imply better design. The axial portion and cube portion were replicated with the center points increased one and three times. The results show that replicating the star points tends to reduce the D and G-optimality criteria of the CCDs (SCCD, RCCD, OCCD and FCCD) at some factor levels while it is different for the A optimality criterion. The results suggest replication affects the different criteria in very different ways because what improves one criterion may be detrimental to a different criterion. The overall results show that G optimality criteria performed better than D and A optimality in different factor levels and replications.

Boosted Regression Tree for Modeling Evaporation Piche Using Other Climatic Factors Over Ilorin


Authors: Ezekiel I. D. ; Alabi N. O.
Pages: 98-106
Abstract
Evaporation is one of the climatic/meteorological factors influenced by causes of climate change. Interest in the topic of climate change has been growing over the last three decades. The threat it poses cannot be overemphasized particularly in developing economies largely due to the connection it has with national development issues. We present a regression tree grown using recursive binary splitting, cost complexity pruning and boosting to study the relationship between evaporation piche and other climatic factors such as relative humidity, solar radiation, sunshine hours, wind speed, temperature and rainfall over the city of Ilorin in Nigeria. These factors are generally seen to change with rising climatic changes in a place. Analysis of the fitted tree reveals that relative humidity, temperature and rainfall are the most important meteorological factors affecting the level of evaporation piche in the city of Ilorin.

A System of Selecting Multi-Index Comprehensive Evaluation Method


Authors: Li Yang ; Wang Jian
Pages: 95-97
Abstract
In this paper, we based on 7 multi index comprehensive evaluation methods, a system of selecting multi-index comprehensive evaluation method is put forward and combined with the multi-index data of air quality in four cities a day, the flow of the multi-index evaluation method proposed above is analyzed, which shows the feasibility of the system and reduces the blindness in the selection of multi-index comprehensive evaluation methods.

Research on the Linkage Effect of Natural Gas Price


Authors: Jian Wang ; Guangshuai Zhou
Pages: 90-94
Abstract
In this paper, we develop a vector error correction model for US natural gas market. It allows us to analyze the linkage effect of the natural gas price from 1998 to 2016. In particular, we prove the evidence that there is a long-term equilibrium relationship in US natural gas, coal and crude oil prices. Impulse response function and variance decomposition are used to examine the linkage effects that a shock in coal and crude oil price would have on natural gas price.