I am a little confused about hypothesis testing. It seems like there are two difference ways to arrive at whether or not to reject or not reject the null hypothesis: calculate the test-statistic OR calculate the confidence interval. In the CFA text, page 460, it alludes to this, and also states that most analysts prefer to use the test-statistic because you can then go on to use the p-value. However, I am not sure what to do with the test-statistic. I’m pretty good on confidence intervals, so just want some input on the test-statistic.
calculate t-statistic if tcalc < tcrit --> then FAIL TO REJECT H0 else Reject H0 tcrit would have to be based on whether one or two tail and Level of significance
Gotcha. Thanks.