A Statistical Analysis of Value of Imports in Nigeria
Nwaigwe,
Chrysogonus Chinagorom,
Iwu Hycinth Chukwudi
Issue:
Volume 3, Issue 5, September 2014
Pages:
117-124
Received:
18 February 2014
Accepted:
7 August 2014
Published:
20 August 2014
Abstract: The value of products imported into a country each year goes a long way to tell us how much the product is appreciated in the country. Some products form the major imports of some countries yearly. Previous studies have focused on the major determinants of imports in some countries, including Nigeria. In the present study, we used a regression approach to identify the major significant imports in Nigeria. This will help in the effective distribution of human resources and services and improve balance of payments. In using the above approach, the method of stepwise regression and transformation of the data via first-order differencing were employed to remove multicollinearity from the data. A predictive model was then specified for the prediction of future imports in the country. Our results show that the major significant imports in Nigeria during the period under study are miscellaneous manufactured goods, machinery and transport equipment, food and live animal, beverages and tobacco.
Abstract: The value of products imported into a country each year goes a long way to tell us how much the product is appreciated in the country. Some products form the major imports of some countries yearly. Previous studies have focused on the major determinants of imports in some countries, including Nigeria. In the present study, we used a regression app...
Show More
Partially Neighbor Balanced Designs for Circular Blocks
Issue:
Volume 3, Issue 5, September 2014
Pages:
125-129
Received:
16 July 2014
Accepted:
14 August 2014
Published:
30 August 2014
Abstract: A partially neighbor balanced design is a design in which for any fixed treatment, other treatments occur as neighbor λi times. This paper generates infinite series of one-dimensional partially neighbor balanced designs for v = n treatments. The blocks used in these designs are considered circular. Designs given here are partially balanced in terms of nearest neighbors and not necessarily in terms of variance. Binary and non-binary concepts have been used for the construction of designs. Theorem 1 generates binary generalized 2-neighbor designs and theorem 2 generates non-binary generalized 3-neighbor designs. These theorems generate designs for v = n treatments i.e., for odd and even number of treatments simultaneously. This concept remains relatively under-explored in the literature. The objective is to decrease error variance due to neighbor effect and reduce computational cost.
Abstract: A partially neighbor balanced design is a design in which for any fixed treatment, other treatments occur as neighbor λi times. This paper generates infinite series of one-dimensional partially neighbor balanced designs for v = n treatments. The blocks used in these designs are considered circular. Designs given here are partially balanced in terms...
Show More
Analysis of Indirect Human Influences and its Bad Impacts on Ecosystems of Natural Forest Resources (Sundarbans) in Bangladesh
Issue:
Volume 3, Issue 5, September 2014
Pages:
130-140
Received:
6 August 2014
Accepted:
18 August 2014
Published:
30 August 2014
Abstract: Sundarbans plays a vital role for human survivability from cradle to grave including tangible and intangible benefits. The total forest area of Bangladesh is about 2.47 million ha, which accounts for about 18% of the total land area of the country, and which constitutes 0.15% of the world’s total tropical forests (Haque, 2000), although an estimate from 1993 only put the tree cover at 5-7% of the country’s area (FAO,1993). Sundarbans comprises 45% of the total productive forests in Bangladesh, and contributes about 50% of forest related revenue (Awal, 2007). The Sundarbans is the largest single mangrove forest in the world, occupying about 6,029 km2 in Bangladesh and the rest in India (Iftekhar & Islam, 2004). At the advent of British rule in 1765, the Sundarbans forests were double their present size (Seidensticker, and Hai, 1983; Khan, 1997). But it is facing tremendous problems (Awal, 207, 2009, 2014). However, a serious killer disease (top dying) of H. fomes in Sundarbans is affecting millions of the trees (Awal, 2007). The loss of H. fomes will have a major impact on the Sundarbans mangrove ecosystem, as well as lead to economic losses (Awal, 207, 2009, 2014). But, it is now under serious threatened through human direct and indirect destruction (Awal, 207, 2009, 2014), and by ecological pollution (Awal, 2007). The cause of this dieback is still unknown (Awal, 2007). The present work has investigated one of the possible factors that might be causing this top-dying namely the concentrations of various chemical elements present in the soil or sediments, particularly, Exchangeable K, heavy metals, though other chemical parameters such as the pH, moisture content of the soil or sediment and nutrient status were also assessed due to indirect human destruction of Sundarbans natural resources (Awal, 2007). A questionnaire survey was conducted among different groups of people inside and outside of Sundarbans to explore local perceptions as to the possible causes of top dying (Awal, 207, 2009, 2014). This confirmed the increase in top-dying prevalence (Awal, 2007).
Abstract: Sundarbans plays a vital role for human survivability from cradle to grave including tangible and intangible benefits. The total forest area of Bangladesh is about 2.47 million ha, which accounts for about 18% of the total land area of the country, and which constitutes 0.15% of the world’s total tropical forests (Haque, 2000), although an estima...
Show More
Investigation of Some Estimators Via Taylor Series Approach and an Application
Tolga Zaman,
Vedat Saglam,
Murat Sagir,
Erdinc Yucesoy,
Mujgan Zobu
Issue:
Volume 3, Issue 5, September 2014
Pages:
141-147
Received:
1 September 2014
Accepted:
12 September 2014
Published:
20 September 2014
Abstract: In this study, the use of taylor series method in the calculation of some means with single auxiliary variable developed in a simple random sampling and mean square error unit ratio estimators having certain properties was investigated. An application was performed in respect thereof. The study population (mass) included 111 secondary schools from 18 districts of Trabzon province. Auxiliary variable (x) was taken as the number of students whereas the main variable (y) was taken as the number of teachers. Sample size was calculated as 45 for unit ratio that has certain features. Afterwards, theoretically proposed mean and units ratio estimators having certain properties were compared numerically. Random sampling was performed using the SPSS 20 program thus giving an equal chance to the units sampled and variability in the population was protected.
Abstract: In this study, the use of taylor series method in the calculation of some means with single auxiliary variable developed in a simple random sampling and mean square error unit ratio estimators having certain properties was investigated. An application was performed in respect thereof. The study population (mass) included 111 secondary schools from ...
Show More
New Criteria of Model Selection and Model Averaging in Linear Regression Models
Magda Mohamed Mohamed Haggag
Issue:
Volume 3, Issue 5, September 2014
Pages:
148-166
Received:
4 September 2014
Accepted:
17 September 2014
Published:
20 October 2014
Abstract: Model selection is an important part of any statistical analysis. Many tools are suggested for selecting the best model including frequentist and Bayesian perspectives. There is often a considerable uncertainty in the selection of a particular model to be the best approximating model. Model selection uncertainty arises when the data are used for both model selection and parameter estimation. Bias in estimators of model parameters often arise when data based selection has been done. Therefore, model averaging of the parameter estimators will be done to alleviate the bias in model selection in a set of candidate models, by combining the information from a set of candidate models. This paper is two-fold, new criteria of model selection are proposed based on different averages of AIC, BIC, AICc, and HQC. Also, model averaging is introduced to compare the parameter estimators in model averaging with the ones in model selection. Two Simulation studies are considered, the first is for model selection and showed that the new proposed criteria are lies between some of the known criteria such as AIC, BIC, AICc, and HQC, and so they can be used as new criteria of model selection. The second simulation study is for model averaging and showed that the parameter estimators have less bias and less predicted mean square error (PMSE) compared with the parameter estimators in model selection.
Abstract: Model selection is an important part of any statistical analysis. Many tools are suggested for selecting the best model including frequentist and Bayesian perspectives. There is often a considerable uncertainty in the selection of a particular model to be the best approximating model. Model selection uncertainty arises when the data are used for bo...
Show More