The data set being used for this report shows the percentage of proein intake from different types of food all over the world. The variables are all different types of foods, an obesity and a COVID-19 percentage based off of cases in the total population. The variable I decided to use for my project was eggs.
In order to find the confidence interval of the mean without bootstrapping, we need to use techniques from more basic statistic classes. We need the sample size (n), standard deviation(s), mean (x), and margin of error (e). We can plug these into the equation to find the confidence interval of the mean. \[ Lower=x-e(x/n) \] \[ Upper=x+e(x/n) \] Using this, we find the condfidence interval of the mean to be
| LCL.95 | UCL.95 |
|---|---|
| 1.043119 | 1.279816 |
The following histogram is showing the bootstrap confidence interval of the means with n=170 to be
| LCL.95 | UCL.95 |
|---|---|
| 1.047809 | 1.278161 |
In conclusion, the confidence interval of the mean is (1.043119, 1.279816) and after using the bootstrap method, the confidence interval of the mean is (1.047809, 1.278161). Since the confidence interval without bootstrapping is wider and it got narrower once we bootstrapped, the bootstrapping method is better to find estimations and make predictions on the percentage of protien intake regarding the eggs variable in the data set.