2 financial analysts were asked to predict earnings per share for a random sampl

Question

2 financial analysts were asked to predict earnings per share for a random sample of 12 corporations over the coming year. The quality of their forecasts was evaluated in terms of absolute percentage forecast error defined as 100 * |Actual –Predicted| / Actual

The absolute percentage forecast errors made are shown in the table below (and are given in worksheet “corp. data” :

CORPORATION ANALYST A ANALYST B
1 12,3 7,3
2 15,4 12,1
3 5,3 7,4
4 9,2 8,1
5 8,6 11,3
6 14,2 12,3
7 5,2 3,1
8 4,1 0,6
9 5,3 5,5
10 4,1 2,8
11 3,6 4,3
12 5,6 1,7

G.1 Which test do you suggest to compare these results? Do this test with all the steps.

Explanation / Answer

Suggest you calculate the mean for each sample, A (predictions from analyst one) and B (predictions from analyst two) The financial analyst who has the lower mean was able to predict earnings per share with less forecast error . Next you need to determine if the difference between the means is SIGNIFICANT. Dependent samples (or "paired") t-tests typically consist of a sample of matched pairs which is what you are looking at here. Dependent t-test for paired samples This is an example of a paired difference test. Select the level of significance: In most of the cases in the paired sample t-test, significance level is 5%, Calculate the parameter: To calculate the parameter use the formula for the paired sample t-test, n=12

Related Questions

Navigate

• Browse (All)
• Browse 2
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.