How to Create a Seaborn Correlation Heatmap in Python | Heatmap | Coorelation Matrix | SuMyPyLab

how to create a Seaborn correlation heat map in our earlier lessons we have learned how to generate heat Maps using python caborn library in this lesson we will learn how to plot correlation Matrix using caborn so what is a correlation Matrix a correlation Matrix is simply a n byn symmetric Matrix that displays the correlation coefficients for different variables large values in the correlation Matrix indicat strong colinearity between the variables involved a correlation Matrix helps us to understand whether a set of variables correlate with each other or not and whether that correlation is positive or negative in this lesson we will learn how to calculate the correlation coefficients for a number of different stock prices by downloading the data from online sources but before that let's understand how to calculate the correlation coefficient of just two variables let's start with a simple data frame with just two variables ccore count and sales here the data is for seven consecutive days let's display the data [Music] frame so these are the customer counts and corresponding sales figures for seven consecutive days let's calculate the correlation coefficient for these two columns of the data frame using Panda's core [Music] function as we can see the correlation coefficient of cust count and sales is 0.969 that is nearly 1 so there exists a high positive correlation between cust count and sales amount let's plot a regression line using map plot lib and Seaborn [Music] Library we will use the seaborn's reg plot function for this we need to pass the data frame and the X and Y variables to the regge plot function let's use cust count as X and sales as y [Music] this is our complete code for calculating the correlation coefficient between cust count and sales and plotting a regression line and here is our regression plot positive correlation is clear from the plot now let's plot a regression line for this new data set here the columns are temp that is temperature and sales here sales figure is for the sales of winter clothes let's modify the code for the change data [Music] set and here is our regression plot notice that the columns have a strong negative correlation I forgot to display the correlation coefficient let's display it [Music] now as we can see the correlation code Co efficient is - [Music] [Music] [Music] see the regression line is flat it indicates that the sales of salt is not dependent on the temperature look here the correlation coefficient is just 0.1 which is nearly zero I hope that you have understood the concept behind the correlation coefficient now let's move to the higher level that is creating a correlation Matrix of more than two variables in this example we will create a data frame from a CSV file the csb f file holds the closing prices of a few stocks listed in various stock exchanges these are the closing prices of the stocks SBN pnb BAC WFC Etc let's start coding again just follow me along while I code I will explain wherever [Music] required creating a data frame DF from the CSV file stock prices. [Music] CSV here we will use the date column as index of the data frame [Music] DF let's have a glance at the data frame so [Music] created [Music] as we can see the data frame has 258 rows and 10 [Music] columns let's drop the rows that have Nan values using the drop in a [Music] function [Music] now the number of rows is reduced to 237 from 258 earlier now let's calculate the correlation coefficients of all the combination of stocks using the co function as a matrix this Matrix is known as a correlation [Music] Matrix [Music] let's display the correlation Matrix coore DF The Matrix is stored as a data frame so we have got a 10x 10 matrix this is because we have 10 stocks in our data set these are the correlation coefficients of each pair of stocks let's enlarge the screen SC and see the Matrix [Music] clearly as we can see the correlation coefficient for pnb and sbin is 0.711 for HDFC and SB it is 0.236 and so [Music] on now let's plot a heat map of this correlation Matrix using Seaborn heat map [Music] function and here is our correlation heat map interesting right so we have successfully plotted a correlation heat map of stock prices from a CSV file using just a few lines of code now we will use the Y finance that is Yahoo finance library for downloading stock prices if this library is not installed in your system you may install it using pip install wi Finance like before just follow me along while I code I will explain the codes wherever necessary we want to download the stock prices for the date range January 1st 2023 to December 31st [Music] 2023 we will use the wif finance. download function for downloading the stock prices for a particular stock the downloaded data can be saved as a data frame as shown here sbns is the ticker name the date range is mentioned using the start and end parameters let's first view the data frame SBI andore DF to check the columns and the [Music] data as we can see the data frame consists of 245 rows and six columns the column names are open high low close adjusted close and volume we need only the adjusted close column so let's download the data for this column only now we have only the adjusted prices for various dates here date is the index column in this way we can download stock prices for any listed stock [Music] now let's create a list of tickers of our interest we will use this list for downloading the adjusted close prices of these tickers in one [Music] go let's modify the code for downloading the prices for all of the tickers as per our list [Music] here instead of a single ticker we have supplied a list of tickers to the download [Music] [Music] function now our data frame stock uncore prices holds the adjusted close prices of the tickers mentioned in the list stock uncore tickers our data frame is ready we can use this data frame for any kind of calculation and [Music] plotting let's modify the list of [Music] tickers now we have 10 tickers let's clean the data frame by dropping the rows with man values [Music] now the data frame has 237 [Music] rows now let's create the correlation Matrix of the stock [Music] prices [Music] and here is our 10x 10 correlation Matrix [Music] let's plot this Matrix as a heat map using [Music] Seaborn and here's our heat map against the correlation Matrix of the adjusted closing prices of the stocks this is very basic we may format this plot with other [Music] parameters now we will modify this code for generating a heat map against the correlation Matrix of returns of each stock we will also format the plot for more beautiful [Music] appearance this time we will plot the correlation Matrix of the daily returns of the [Music] stocks [Music] let's plot the heat map with a few other [Music] parameters let's assign the axis object to the variable [Music] ax now let's give a title to the [Music] plot increasing the font size for the [Music] plot [Music] set the background style as [Music] white let's format the numbers with two decimal [Music] [Music] places and here is our final heat map isn't it beautiful you may format this plot with many other ways for making it more [Music] beautiful here is our complete code [Music] hope that you have enjoyed this lesson and learned many new things if you find this lesson helpful please don't forget to like share comment and subscribe your valuable feedback is important for me and will help me improve the contents of the channel thanks for watching up to the end hope to see you in the next bye

Channel: SuMyPyLab

Views: 662

Keywords: How to Create a Seaborn Correlation Heatmap in Python, Coorelation Matrix, Heatmap, Seaborn Correlation Matrix, Seaborn Correlation Heatmap, Heatmaps in Seaborn, Heatmaps in Python, Data Science, Data Analytics, Data Analysis, Visualization in Python, Data Visualization in Python, Heatmap Visualization in python, correlation matrix visualization in python

Id: K3zlMLQAZlw

Length: 16min 33sec (993 seconds)

Published: Fri Jan 05 2024