We started the class with the sigma rules of normal distribution (see the first workbook of the above classroom task).
Then we reviewed classroom notes II of the third class. After that we understood why the "unbiased variance" is unbiased, i.e., we understood why it is better to divide with (n-1) instead of n. Then we learned that the standard error of a statistic is the data based estimation of its standard deviation. We learned the "standard error of mean" statistic and we took a glimpse into the content of the above "skewness and kurtosis" link. Then we calculated the basic statistics of the variables day 2 cholesterol and day 4 cholesterol. We answered the question of the physician (see the above blackboard photo): also taking into account the standard error of mean statistic (approximately 9) the observed mean (approximately 254) does not contradict the statement of the scientific paper.
Then we created the empirical distribution function of day 2 cholesterol. For technical reason we replaced the jumps with line segments. Finally, we supplemented the chart with the distribution function of the best fitting normal distribution (the expectation and the standard deviation were estimated from the data).
