Statistik Deskriptif dan Inferensial

Menganalisis Data Survei di Python

EbunOluwa Andrew

Data Scientist

Statistik deskriptif

Ukuran dasar untuk mendeskripsikan data survei
Contoh: mean, median, modus, rentang, simpangan baku, dll.
.describe()

Foto oleh Lukas - grafik jarak dekat

¹ Photo by Lukas

Fungsi .describe()

data.describe()

|      | year     | satisfaction_rating
|------|----------|--------------------
| count|       42 |                  42
| mean | 2012.381 |            7192.857
| std  |    4.196 |             945.178
| min  |     2006 |                5500
| 25%  |     2009 |                6325
| 50%  |   2012.5 |                7400
| 75%  |     2016 |                8000
| max  |     2019 |                8600

data.describe(include = np.object)

|        | category    |
|--------|-------------|
| count  | 42          |
| unique | 3           |
| top    | Residential |
| freq   | 14          |

Menafsirkan .describe()

Pencilan = nilai maks > nilai mean & median
Nilai tidak masuk akal = jika nilainya tidak logis

|       | year     | satisfaction_rating |
|-------|----------|---------------------|
| count |       42 |                  42 |
| mean  | 2012.381 |            7192.857 |
| std   |    4.196 |             945.178 |
| min   |     2006 |                5500 |
| 25%   |     2009 |                6325 |
| 50%   |   2012.5 |                7400 |
| 75%   |     2016 |                8000 |
| max   |     2019 |                8600 |

Menafsirkan .describe()

Top = modus = kelas paling sering muncul
Freq = jumlah kemunculan kelas tertinggi

|        | category    |
|--------|-------------|
| count  | 42          |
| unique | 3           |
| top    | Residential |
| freq   | 14          |

.describe() pada electric_satisfaction

import pandas as pd

electric_satisfaction = pd.read_csv("austin-energy-customer-satisfaction.csv")

.describe() pada electric_satisfaction

electric_satisfaction.describe()

|      | year     | satisfaction_rating
|------|----------|--------------------
| count|       42 |                  42
| mean | 2012.381 |            7192.857
| std  |    4.196 |             945.178
| min  |     2006 |                5500
| 25%  |     2009 |                6325
| 50%  |   2012.5 |                7400
| 75%  |     2016 |                8000
| max  |     2019 |                8600

satisfaction_rating memiliki pencilan
Persentil ke-50 = median

.describe() pada electric_satisfaction

|        | category    |
|--------|-------------|
| count  | 42          |
| unique | 3           |
| top    | Residential |
| freq   | 14          |

Modus = responden residensial

Statistik inferensial

Menilai apakah data dapat digeneralisasi ke populasi lebih luas
Ukuran sampel < ukuran populasi -> galat sampling
Mengestimasi parameter populasi
- Interval kepercayaan
  - fungsi norm.interval()

Foto oleh Andrea Piacquadio di Pexels - perempuan memegang bohlam

¹ Photo by Andrea Piacquadio on Pexels

Fungsi norm.interval()

Untuk dataset besar
Asumsikan sebaran sampling mean berdistribusi normal

import scipy.stats
scipy.stats.norm.interval(alpha, loc, scale)

alpha = tingkat kepercayaan
loc = mean sampel
scale= galat baku sampel

Menafsirkan norm.interval() pada electric_satisfaction

electric_satisfaction = pd.read_csv(
  "austin-energy-customer-satisfaction.csv")

conf_interval = st.norm.interval(
  alpha = 0.99,
  loc = np.mean(electric_satisfaction.satisfaction),
  scale=st.sem(electric_satisfaction.satisfaction))


print(conf_interval)

(6817.187361704269, 7568.526924010017)

Ayo berlatih!

Menganalisis Data Survei di Python