In: Statistics and Probability
Once again, as has happened in the past, you are very much in doubt concerning the validity of the known population standard deviations, this time for each camera body, in the yearly sales of the two brands of camera bodies. Therefore, you wish to conduct your study with the knowledge that the population standard deviations are unknown. You collect random samples of the yearly sales of the two camera bodies at populations of stores. The data that has been collected is shown in appendix one below. At both the 10% and 5% levels of significance, are there any differences in the mean sales of the two camera bodies at the two populations of stores? Again, if the software makes it possible, find both 90% and 95% confidence intervals for the difference in the mean sales of the camera bodies between the two populations of stores. Explain the meanings of these intervals. Then, if possible, based upon the procedures you have chosen to address the problem, use the intervals to supplement and test whether there is a difference in the mean sales of the two camera bodies between the two populations of stores.
Appendix One: (Sales of Camera Bodies)
Nikon D5:
131 145 150 156 176 154 138 122 130 235 165 168 221 229 154 155 154 160 154 144 240 143 232 238 130
Canon Model:
138 140 237 147 170 155 232 228 135 130 161 160 220 229 155 158 150 250 248 246 139 233 133 230 126
Appendix Two: (Includes the Purchase of a Lens? Y = yes, N = no)
Nikon D5:
Y N N N Y Y N Y N Y Y N N N N Y Y Y N Y Y N N N N N N Y N N N Y N N N N
Canon Model:
N N Y Y Y N N Y N Y N N Y Y Y Y N N N N Y N Y N Y N N N Y Y Y Y Y N Y Y
We are given the following data and are asked to test the hypothesis that there is a difference in the mean sales of both the camera bodies and also find the confidence interval at 90% and 95% confidence.
Nikon D5:
131 145 150 156 176 154 138 122 130 235 165 168 221 229 154 155 154 160 154 144 240 143 232 238 130
Canon Model:
138 140 237 147 170 155 232 228 135 130 161 160 220 229 155 158 150 250 248 246 139 233 133 230 126.
We now have to calculate mean and standard deviation for the two camera bodies.
Let Nikon D5 be denoted by x and Canon Model be denoted by y
xi | xi2 | yi | yi2 |
---|---|---|---|
131 | 17161 | 138 | 19044 |
145 | 21025 | 140 | 19600 |
150 | 22500 | 237 | 56169 |
156 | 24336 | 147 | 21609 |
176 | 30976 | 170 | 28900 |
154 | 23716 | 155 | 24025 |
138 | 19044 | 232 | 53824 |
122 | 14884 | 228 | 51984 |
130 | 16900 | 135 | 18225 |
235 | 55225 | 130 | 16900 |
165 | 27225 | 161 | 25921 |
168 | 28224 | 160 | 25600 |
221 | 48841 | 220 | 48400 |
229 | 52441 | 229 | 52441 |
154 | 23716 | 155 | 24025 |
155 | 24025 | 158 | 24964 |
154 | 23716 | 150 | 22500 |
160 | 25600 | 250 | 62500 |
154 | 23716 | 248 | 61504 |
144 | 20736 | 246 | 60516 |
240 | 57600 | 139 | 19321 |
143 | 20449 | 233 | 54289 |
232 | 53824 | 133 | 17689 |
238 | 56644 | 230 | 52900 |
130 | 16900 | 126 | 15876 |
Since population standard deviations are not known, we use pooled standard deviation.
Null hypothesis
i.e., there is no difference in the mean sales of both the camera parts
Alternate hypothesis
i.e., there is a difference in the mean sales of both the camera parts
Test statistic under H0:
Inference
at 5% significance level, 48 degrees of freedom and for two-tailed test from t-table
=> We fail to reject the null hypothesis.
Hence we conclude that there is no difference in the mean sales of the two camera bodies.
at 10% significance level, 48 degrees of freedom and for two-tailed test from t-table
=> We fail to reject the null hypothesis.
Hence we conclude that there is no difference in the mean sales of the two camera bodies.
The formula for calculating 90% confidence interval for the difference in the mean sales of the two camera bodies is,
=>
=>
Hence the 90% confidence interval is [-4.04,30.12]
This means that we can be 90% confident that the true mean difference lies between -4.04 and 30.12.
For 95% confidence only the critical value will change, the formula remains the same.
=>
Hence the 95% confidence interval is [-7.40, 33.48]
This means that we can be 95% confident that the true mean difference lies between -7.40 and 33.48.
Note:- I made a mistake here, I took the smaller mean as the first mean i.e., x-bar should have been mean of Canon Model and y-bar should have been mean of Nikon D5, this does not change the result of the hypothesis test but it could have affected our confidence intervals. So I interchanged the means so that we could get the correct intervals but the rest remains the same.