Statistical sampling is a powerful tool used to infer characteristics of a population based on a subset, or sample, of the population. As you can well imagine, selecting an appropriate sample and then inferring the results back to the population from which it came is critically important in business.
Data should be gathered from a sample that resembles the population of interest. Once these data have been gathered, sample statistics can be applied to solve business problems. The sample statistics that are most helpful to business professionals include mean, variance, and standard deviation. Sample statistics, along with appropriate probability distributions such as Z and t-distributions, can be used to compute confidence intervals. With a 95 percent confidence interval, business managers can proceed in their decision making with a level of assurance that comes from knowing the population itself.
While statistical analysis enables business managers to make inferences about the current state of the population, regression enables decision makers to look at a data sample and make predictions. Regression equations depend on the covariance and correlation of two variables. The correlation of two variables is also known as the correlation coefficient. Covariance and correlation compare the movements of two random variables. The regression equation uses covariance to describe the relationship between two or more variables and, based on the past performance of these variables, predict future behavior. Regressions are used to forecast sales and project market movements for stocks and bonds.
When you have completed "Statistical Sampling and Regression," you should be able to
- describe the characteristics of an unbiased sample
- calculate a sample's measures of central tendency and measures of dispersion
- develop a confidence interval around a sample mean for large and small samples
- calculate covariance and correlation
- understand the application for and interpret a simple linear regression