Intervals at three confidence levels

From raw data, build confidence intervals at three levels and see them widen.

What to build

A program that prints a mean, its standard error and three confidence intervals.

Requirements

  • Print mean M (2 decimal places), standard error S (3 decimal places), then 90%: L to U, 95%: L to U and 99%: L to U (2 decimal places).

Input

One line: numbers separated by spaces.

Output

Five lines.

Constraints

2 to 1000 numbers.

Example

Input:

162 170 168 175 159 181 166 172 169 177 164 171 174 160 168

Output:

mean 169.07
standard error 1.617
90%: 166.41 to 171.73
95%: 165.90 to 172.24
99%: 164.90 to 173.23

Why it works

Each level uses the z value that leaves that share of a normal curve in the middle, so more confidence multiplies the same standard error by a larger number.

Examples

Example 1 · Input
162 170 168 175 159 181 166 172 169 177 164 171 174 160 168
Expected output
mean 169.07
standard error 1.617
90%: 166.41 to 171.73
95%: 165.90 to 172.24
99%: 164.90 to 173.23
Example 2 · Input
12 15 11 18 14 16 13 17
Expected output
mean 14.50
standard error 0.866
90%: 13.08 to 15.92
95%: 12.80 to 16.20
99%: 12.27 to 16.73

Prompts inside input() are ignored when checking, so input("n: ") is fine.

Hints

Hint 1

Use statistics.stdev for the sample standard deviation.

Hint 2

The z values are 1.645 for 90%, 1.96 for 95% and 2.576 for 99%.

Solution

Show one correct solution (try it yourself first)
import statistics as st

data = list(map(float, input().split()))
m = st.mean(data)
se = st.stdev(data) / len(data) ** 0.5
print(f"mean {m:.2f}")
print(f"standard error {se:.3f}")
for level, z in (("90%", 1.645), ("95%", 1.96), ("99%", 2.576)):
    print(f"{level}: {m - z * se:.2f} to {m + z * se:.2f}")

Lesson: Confidence intervals →