Descriptive Stats.ipynb

Kernel: Python 3

import numpy as np 
A=[1,2,3,4,5,6] ##- List 
A+A
D = np.array(A)
D

D.sum()

D.mean()

np.std(D)

np.median(D)

np.var(D)

np.std(D[0:3]) # sd for 2 at 2. 

a=[[1,2,13,0],[4,5,6,78],[-8,8,-6,9]]
z=np.array(a)
z

z[1]

z[1,1]

z[:,2]

# Sum of elements from 3rd row 
# mean of elements of 4th column
# std deviatian for array
# median across 2th row 
# mean across 1 and 3 row 

(D[0].mean(),D[2].mean())

3*7

Analyzing Car sales Data

import numpy as np                                                 # Implemennts milti-dimensional array and matrices
import pandas as pd                                                # For data manipulation and analysis
#import pandas_profiling
import matplotlib.pyplot as plt                                    # Plotting library for Python programming language and it's numerical mathematics extension NumPy
import seaborn as sns                                              # Provides a high level interface for drawing attractive and informative statistical graphics
%matplotlib inline
sns.set()

from subprocess import check_output

Importing Data

s=[1,2,3]
pd.Series(s)

pd.DataFrame(s)

Car= pd.read_excel("Car_Sales.xlsx")# data set is located in the same folder
#Car=pd.ExcelFile(r"C:\Users\suyashi144893\Documents\data Sets\Car_Sales.xlsx").parse("Sheet1")

Car.shape

Car.head() 

Car.tail(2)

Car.describe()

Car.describe(include="all")

Car.isnull().sum()

a=Car["engV"].mean()
a

b=Car["engV"].median()
b

Car["engV"]= Car["engV"].fillna(b)

Car=Car.dropna()

Car.isnull().sum()

Car.duplicated().sum()

Car.iloc[:,1]

Car.loc[Car.duplicated(keep=False),:]

Car.duplicated().sum()

Car.drop_duplicates(keep='first').shape

Visulaizations

Car["mileage"].plot.hist(bins=10)

sns.countplot(x="body",data=Car).set_title("Count plot for Car variants")

Car.body.count()

pd.crosstab(index=Car["body"],columns="count")

Price distribution between registered and non-registered cars

sns.boxplot(x='registration',y='price',data=Car)

Establish coorelation between all the features using heatmap.

corr = Car.corr()
plt.figure(figsize=(10,10))
sns.heatmap(corr,vmax=.8,linewidth=.01, square = True, annot = True,cmap='YlGnBu',linecolor ='black')
plt.title('Correlation between features')

Car.groupby(['engType', 'drive']).count()['body']

Engtype VS Drive

Car[Car['engType'] == 'Petrol'].drive.groupby(Car.drive).count().plot(kind='pie', figsize=(6, 6),autopct='%1.2f%%')
plt.axis('equal')

sns.barplot(x='registration', y='price', data=Car)

sns.swarmplot(x='registration', y='price', data=Car)