Summary  
This chapter covers how to implement code that computes a 95% confidence interval by calculating the sample mean, sample standard deviation, standard error, determining the critical z-score, margin of error, and then printing the lower and upper interval bounds.

General domain of usage  
Statistical data analysis

import unittest
import pandas as pd
import numpy as np
import scipy.stats as st

def _dynamic_test(test_case, condition, success_msg, failure_msg):
    if condition:
        test_case._testMethodName = success_msg
        test_case.assertTrue(True, success_msg)
    else:
        test_case._testMethodName = failure_msg
        test_case.fail(failure_msg)


class TestConfidenceInterval(unittest.TestCase):
    def test_confidence_interval(self):
        import user_code

        # ÐÐ°Ð²Ð°Ð½ÑÐ°Ð¶ÑÑÐ¼Ð¾ ÑÑ ÑÐ°Ð¼Ñ Ð´Ð°Ð½Ñ
        data = pd.read_csv(
            'https://codefinity-content-media.s3.eu-west-1.amazonaws.com/a849660e-ddfa-4033-80a6-94a1b7772e23/section_1/mean_mass.csv'
        )

        # ÐÑÑÐºÑÐ²Ð°Ð½Ñ ÑÐµÐ·ÑÐ»ÑÑÐ°ÑÐ¸
        mean_list = data['mass']
        expected_confidence = st.norm.interval(
            confidence=0.95,
            loc=np.mean(mean_list),
            scale=st.sem(mean_list)
        )

        # ÐÐµÑÐµÐ²ÑÑÐºÐ° ÑÐ¸Ð¿ÑÐ² Ñ ÑÐ¾ÑÐ½Ð¾ÑÑÑ
        condition = (
            isinstance(user_code.mean_list, pd.Series)
            and isinstance(user_code.confidence, tuple)
            and abs(user_code.confidence[0] - expected_confidence[0]) < 1e-6
            and abs(user_code.confidence[1] - expected_confidence[1]) < 1e-6
        )

        _dynamic_test(
            self,
            condition,
            "The confidence interval is calculated correctly.",
            "The confidence interval is incorrect. Check your calculation parameters."
        )


if __name__ == "__main__":
    unittest.main()


test_code.py

Pythonを用いたデータ分析で使用される主要な統計概念を学習します。本コースでは、平均、中央値、最頻値、分散、標準偏差などの記述統計に加え、サンプリング、確率分布、中心極限定理、外れ値検出についても取り上げます。

95%信頼区間の計算

解答