Before calculating mean, median and mode, let us look at types of data and characteristics of the data. At a very high level data can be classified as categorical and quantitative data. Both can be further classified as below
| Difference | Order | Similar Interval | Meaningful Zero | ||
| Categorical | Nominal (Cities) | Yes | – | – | – |
| Categorical | Ordinal (Temp.) | Yes | Yes | – | – |
| Quantitative | Interval | Yes | Yes | Yes | – |
| Quantitative | Ration | Yes | Yes | Yes | Yes |
Now all of these types of data do not have all characteritics
| Mode | Median | Mean | |
| Nominal | Yes | – | – |
| Ordinal | Yes | – | – |
| interval | Yes | Yes | Yes |
| ratio | Yes | Yes | Yes |
Mean:
Mean is nothing but average. It can be calculated in python or by using numpy
Median
Middle value of observation when ordered from low to high
Mode
Mots commonly occurring observation
| ################################################################################################ | |
| # name: discriptive_statistics_01.py | |
| # desc: identify type of progression | |
| # date: 2018-12-22 | |
| # Author: conquistadorjd | |
| ################################################################################################ | |
| import numpy as np | |
| from scipy import stats | |
| #Calculate mean by python | |
| input_data = input('Input elements separated by comma :') | |
| # Convert input into List | |
| input_list = list(map(int, input_data.split(','))) | |
| print ("input_list", input_list , type(input_list)) | |
| # Mean calculation using simple python | |
| mean = sum(input_list)/ len(input_list) | |
| print('mean', mean) | |
| # Mean calculation using numpy | |
| mean = np.mean(input_list) | |
| print('mean', mean) | |
| # Median calculation using numpy | |
| median = np.median(input_list) | |
| print('median', median) | |
| # Mode calculation using scipy | |
| mode = stats.c(input_list) | |
| print('mode', mode) |