Question

He simulation results of two algorithms in an engineering lab show that 50 error messages in a sample of 500 runs of algorithm versus 15 error messages in a sample of 200 runs of algorithm2. An engineer wants to test if the proportion of the error messages of algorithm is higher than that of algorithm2. Use a-0.05 in your test.

write the letter as a, b, c
1- Is this (8: one-talled or 1b: two-tailed) test?
2- The alterative hypothesis is Mi(p1-p20, b: p1-p20, cp1 = p2 not equal 0)
3- The ortical value is :
4-The test statistic is (round to two decimal places)
5-Since the test statistic is more bless)_________than the critical value, we (a: dob: do not reject) the null hypothesis.

Answer 1

This is a one-tailed test to compare the proportion of error messages in two algorithms. The test statistic is calculated using the formula for two population proportions.

Step-by-step explanation:

This is a one-tailed test because the engineer wants to test if the proportion of error messages of algorithm is higher than that of algorithm2.

The alternative hypothesis is represented as: Ha: p1 > p2, where p1 is the proportion of error messages in algorithm and p2 is the proportion of error messages in algorithm2.

The critical value for the test is not provided in the given information.

The test statistic can be calculated using the formula for two population proportions: z = (p1 - p2) / sqrt((p1(1 - p1) / n1) + (p2(1 - p2) / n2)), where n1 and n2 are the sample sizes for algorithm and algorithm2 respectively.

Finally, since the test statistic is more bless than the critical value, we do not reject the null hypothesis at a significance level of 0.05.