CFA Practice Question
For a one-tailed hypothesis test, the critical value of the test statistic is -2.33. Which of the following is true about the hypothesis test?
A. alpha = .01 for an upper-tailed test
B. alpha = .05 for an upper-tailed test
C. alpha = .01 for a lower-tailed test
Explanation: This is a one tailed test and the z-value of -2.33 suggests that it is a lower tailed test and that it is for a 1% level of significance.
User Contributed Comments 6
| User | Comment |
|---|---|
| mattg | -2.33 test statistic, obviously we are testing the lower tail here, you know it is C right off |
| czar | thanks maatg, hadnt caught that |
| tommyguard3 | Good pooint mattg. On the exam will they provide the Z chart or are we actually supposed to know them? |
| GBolt93 | I believe you have to know the statistic for 68, 95, 99.8 and be able to logic from there. e.g. it's a one tailed test so the test statistic for .01 is between 2 and 3 which is the statistic for 95 and 99.8 for two tailed tests. |
| CalebMast | I had to relearn the tables slightly. First of all, critical value must correspond to z value or -2.33 assumes only means z, not t table. This was the first concern - primary indicator here is that there are no # for population. For unknown variance (unknown population in this case, as well), we must go with the z table. Confirming z not t was the first concern. The next thing was reading the table properly. I was a little sloppy on this, so I got a consultant. This table was helpful: www.intmath.com/counting-probability/z-table.php Once I remembered the brilliant fact that the decimal partials at the top of the graph complete the #'s at the left, it came together perfectly. The value at -2.33 is .0099 (.01). Speaking to what I mentioned above and a concern mentioned by others, everything lower-tail is negative, whereas the positives are upper-tail. |
| Xuengzilam | A one-tailed test evaluates whether a parameter is either greater than or less than a hypothesized value (but not both). There are: Upper - tailed test: You reject the null if the test statistic is too high. Lower-tailed test: You reject the null if the test statistic is too low. the question told you: The rejection region is in the left tail (negative side of the z-distribution). This is clearly a lower-tailed test. In standard normal distribution, z = -2.33 corresponds to the 1st percentile. |