Estimation and Hypothesis Testing

Abstract

This chapter presents basic elements of parameter estimation and hypothesis testing. The reader will learn how to form confidence intervals for the mean, and more generally, how to calculate confidence intervals for the one parameter setting and for the difference between two groups. Principles of hypothesis testing are detailed, including the choice of the null and alternative hypotheses, the significance level, and implications for choosing a one-sided versus two-sided test. The p-value is defined and a discussion of controversies that have arisen over its use are included. After reading this chapter, the reader will have a better understanding of the necessary steps to set up a hypothesis test and make valid inference about the quantity of interest. Other topics in this chapter include exact hypothesis tests, which may be preferable for small sample settings, and the choice of a parametric versus nonparametric test. The chapter also includes a brief discussion of the implications of multiple comparisons on hypothesis testing and considerations of hypothesis testing in the setting of noninferiority trials.