Name:_________________
Confidence Intervals simulation demo
Generate random data distributions
https://davidwills.us/math103/distro_generator.html
Generate a UNIFORM distribution from 1 to 100 of 100000 continuous numbers of 2 decimal digits.
This will be the "population".
Copy and paste it into:
*CI sim demo
https://davidwills.us/stat200/CI_simulation.html
N=________
μ=________
σ=________
sample size n= 10 1000 samples
Confidence level=95% % of Confidence Intervals that contain μ=________
tc=_______
Mean x̄=_______ Mean s=_______ Mean SEM=_______ Mean E=_______
Confidence level=99% % of Confidence Intervals that contain μ=________
tc=_______
Mean x̄=_______ Mean s=_______ Mean SEM=_______ Mean E=_______
Even for a very small sample size of 10 is the percentage of the confidence
intervals that contain μ close to the confidence level:_____
sample size n= 100 1000 samples
Confidence level=95% % of Confidence Intervals that contain μ=________
tc=_______
Mean x̄=_______ Mean s=_______ Mean SEM=_______ Mean E=_______
Confidence level=99% % of Confidence Intervals that contain μ=________
tc=_______
Mean x̄=_______ Mean s=_______ Mean SEM=_______ Mean E=_______
Larger sample size n makes the tc ________
and the (mean) SEM and (mean) E much _________.
Based on this simulation experiment, does the confidence level indicate
the percentage of confidence intervals that would contain μ? _____
NB. this will work even better if the population is normal
(everything is better if it's normal).
NB. if the population is right-skewed (try Exponential λ=2)
the percentages will fall short a few percents.
Try them. It will take only a couple of minutes.