University of Phoenix
Managerial Decision Making- MBA/510
March 7, 2007
Applying parametric and nonparametric statistical techniques
The coffee time simulation and applying parametric and nonparametric statistical techniques taught me to know the difference between these two tests and what they are based of. These two tests are used to compare measurements. The parametric test is based on assumptions on the population distribution. However nonparametric does not make assumptions; this test ranks the outcome variable from low to high and analyzing the ranks. For example, when doing the coffee time simulation I learned that when there is a high degree of correlation between two variables it is a good idea to remove one. In the simulation I chose to remove the value X3: 0.657 since it is the one correlated with greater number of independent variables in the model. The reason why this value had to be removed is because it distorts the value of the regression coefficients. Also, another lesson learned from the simulation was if choosing the parametric test, the data is to be sampled from a population. Nonparametric tests are used for ranking, scores and scale. In the simulation I was playing around on selecting the highest percent and sample size just to see what would happen and my selection was approved to be the correct choice. By selecting the larger sample size to the selected alpha detect even small differences between sample statistics and population parameter.
As a result of using the coffee time simulation I will be able to apply the following concepts and tools at my workplace; multiple regression, z test for proportions and the chi-square test. For example, when having to do a project the multiple regression will help me explain the d ...