In: Statistics and Probability
The 2010 insurance premiums from both Geico and 21st Century for
a male driver licensed for 6–8 years who drives a Honda Accord
12,600 to 15,000 miles per year and has no violations are given in
the table below. The table also indicates the city of residence.
City Geico ($) 21st Century ($) Difference ($) Long Beach 2780 2352
Pomona 2411 2462 San Bernardino 2261 2284 Moreno Valley 2263
2520
(a) Fill in the entires in the difference column in the above table.
(b) Compute the mean of the differences. (c) Verify that the sample
standard deviation of the differences is s ≈288.68. (d) Compute the
standard error of the differences, using n =4. (e) Using a
significance level of α =0.01, do these data provide sufficient
evidence to indicate that there is a difference in the premiums
between Geico and 21st Century?
(a)
City Geico ($) | 21st Century ($) | Difference ($) | |
Long Beach | 2780 | 2352 | 428 |
Pomona | 2411 | 2462 | -51 |
San Bernardino | 2261 | 2284 | -23 |
Moreno Valley | 2263 | 2520 | -257 |
(b)
Mean of the difference is,
(c)
Sample standard deviation of the difference is,
(d)
Standard error of the difference is,
(e)
The hypothesis are,
Ho : There is no difference in the premiums between Geico and 21st century.
Ha : There is difference in the premiums between Geico and 21st century.
The test statistics is,
t = Mean difference / Standard error of difference
t = 24.25/144.34
t = 0.168
Here, = 0.01 and df = n - 1 = 3
Using table of critical values for t-distribution the critical values are -5.841 and 5.841
Since test statistics value does not falls in the critical region, we fail to reject Ho.
Therefore, there is no sufficient evidence to indicate that there is difference in the premiums between Geico and 21st century.