Question

Confidence intervals can be used to indicate the variation in data as well as to determine how many different random simulations are necessary to ensure that the simulation error is withingiven bounds with high probability

a) To bound the expected relative error that the simulation (i.e. the specific data set) introduces on the results of a given experiment, we want to determine the number of simulations we should run. For this, we first run 30 experiments, leading to data with a mean of 4.5 and a standard deviation of 2. Determine whether this data already provides an expected relative error inCntervals can be used to indicate the variation in data as well as to determine how many different random simulations are necessary to ensure that the simulation error is withingiven bounds with high probability

a) To bound the expected relative error that the simulation (i.e. the specific data set) introduces on the results of a given experiment, we want to determine the number of simulations we should run. For this, we first run 30 experiments, leading to data with a mean of 4.5 and a standard deviation of 2. Determine whether this data already provides an expected relative error in