University of Regina Open Journals
Not a member yet
369 research outputs found
Sort by
Analysis of r×c Tables containing Outliers
No full textWe examine the problem of accommodating outliers in tables of measurement data when there are no replications. The suggested strategy for the determination of the cells that contain outliers and for handling them analytically is easily understood and implemented in practice and presented in even the most elementary statistics course along with the course’s ANOVA material
An Inflated Model Based on Sized Biased Himanshu Distribution and its Applications
No full textThe purpose of this paper is to develop an inflated probability model through size biased Himanshu Distribution under the certain assumptions. The proposed model involved certain parameters which are estimated by the Method of Maximum likelihood for the valuable inferences. The suitability of the proposed Model is tested by fitting to observed set of data
A Note on Early Termination in the 2^(k-p) Designs
No full textA complete factorial experiment can resolve problems that occur in experiments that to determine suitable tolerances for the factors in a manufacturing process. Two-level fractional factorial designs have been widely used to investigate the effect of factors in which several factors are involved. In many circumstances, the experiment may be stopped before all the points have been run. This article argues that one should not arrange the points of the experiment in random order. Instead, one should consider adjusting the run order to protect against the risk of early termination, that is, a systematic run order should be carefully applied to the case. In this article, we will use semi-folding design as examples to illustrate the idea of how to take advantage of the sequential run order
A Candy Lover Learns to Optimize
No full textA container has two types of candies: Type A and Type B. Concerned about her child\u27s well-being, a wise mom pronounces, “Each day, you can choose two candies from the container random. If they are of different types, you can eat them both. If they are of the same type, eat only one and return the other to the container.” We analyze the expected number of days needed to eat all candies in the container and the proportion of days the child eats two candies. Several other variations are either worked out or left for readers to solve
Noncalculus Derivations of Least Squares Estimates
No full textThis letter responds to and reflects on goals and claims of the previous issue’s papers on deriving least-squares estimates without calculus and brings in the technique of translating data points. 
The Coefficient of Dependence and Conditioning
No full textThe Coefficient of Dependence is introduced as an avenue for aiding student understanding of dependent, statistical events. The essential features of this coefficient are studied using different models and explored with multiple examples. Conditional probabilities are ultimately understood to be simple transformations of marginal probabilities via the Coefficient of Dependence, which is an idea well within the grasp of all undergraduate students. For completeness, proofs for the main results are relegated to the appendix for those interested
A Review of Kaandossiwin: How We Come to Know Indigenous Re-Search Methodologies (2nd ed.)
No full textA Review of Absolon, K. E. (2022). Kaandossiwin: How we come to know Indigenous re-search methodologies (2nd ed.). Fernwoo
Statistical Modeling for Flood Frequency in Upper Chaophraya River Basin
No full textThe research deals with the probabilistic estimates of annual maximum flood peaks in the upper Chaophraya basin (Thailand) used the extreme value theory, the Block Maxima. The Generalized Extreme Value (GEV) distribution model, the Frèchet distribution (EV2), the extension of the Frèchet distribution such as the Kumaraswamy Frèchet distribution and the new distribution as called the Exponentiated Kumaraswamy Frèchet distribution satisfied the Goodness of fit test (Kolmokorov-smirnov test). The return levels are estimated for 3, 5, 10, 30, 50, 100, 500 and 1000 years which are consistently increasing for designs of flood protection in future. The return period of flood for each stations are estimated.
The investigation of the new distribution and appropriated estimation technique for the flood frequency in upper Chaophraya river basin as we called the Exponentiated Kumaraswamy Frèchet distribution and the differential evolution maximum likelihood estimation were done. We derived the properties of the Frèchet family; such as the Frèchet distribution (EV2), the Kumaraswamy Frèchet distribution (KF) and the Exponentiated Kumaraswamy Frèchet distribution (EKF). We also compared Bias, Variance, Mean Square Error and Mean Absolute Percentage Error for all parameters in each distribution by generating the Frèchet family random number. For the effectiveness of analytical solutions of the parameters we provided the numerical solutions (differential evolution method) to obtain estimates for all parameters by using Scilab program.
Accuracy of flood assessment of extreme event is of fundamental importance for many safety, engineering and financial application. In part of application we provided the probabilistic estimates of annual maximum flood peaks or momentary peak data in the upper Chaophraya river basin (Thailand). The Generalized Extreme Value (GEV) distribution model were used to be gain to compare with as the Frèchet distribution (EV2), the Kumaraswamy Frèchet distribution (KF) and the Exponentiated Kumaraswamy Frèchet distribution (EKF). The Goodness of fit test, the return level and return period were done. The return periods of flood were classified by hazard class using GEV found that in upper Chaophraya river basin flood occurred highly. The result from the Frèchet family also occurred highly, but the return period and return level from the Kumaraswamy Frèchet distribution (KF) quite closed to GEV more than another distribution
Transmuted Modified Weibull Distribution for Modeling Skewed Lifetime Dataset; Properties and Application
No full textA new five parameter model called Generalized Modified Weibull distribution is proposed and studied. The new distribution generalizes the Modified Weibull Distribution introduced by Lai et al. (2003) using transformed – Transformer framework of Alzaatreh et al. (2014) which resulted to a distribution capable of modeling skewed data (positively or negatively skewed data), as well as, symmetric data. Property of a proper probability density function was used to ascertain that the resulting function is a proper probability density function. Statistical properties of the newly generated distribution were studied. Graph of probability density function of the distribution was used to show that it is capable of modeling skewed data. Graph of cumulative density functions of the distribution was plotted using varying parameter values. Monte Carlo simulation approach was used for the test of homogeneity of the distribution and it was observed that the parameters in the distribution approach the true values as sample size increases. Maximum likelihood Estimation method was used for estimation of the model parameters. Real life dataset was used for model comparison as well as demonstration of its application