Algorithmic trading in Python is a game-changer for quantitative traders and crypto trading strategies. This video dives deep into Pandas, one of the most essential libraries for data analysis and manipulation. Since 70% of algorithmic trading involves working with data, mastering Pandas is crucial. From handling time-series data to visualizing 3D candlestick charts, this tutorial covers it all. Whether you’re in the USA using the best algorithmic trading software or exploring quantitative analysis for trading in Singapore, this guide will help you elevate your Freqtrade strategy.
If you are good in data then everything becomes so easy corre so let me quickly show you that what you can achieve with the fanda you can quickly create these kind of extraordinary plots with the pandas let me show you a few more and if you work hard then you can also learn how to create 3D Candlestick charts with volume bars right this is just a simple trailer it depends on your hard work it depends on your imagination and it depends on your consistency that how you can use the pandas in your algo trading so let me show you Candlestick chart you can see here it is really awesome and beautiful and this is what you see in all the exchanges right you can even create 3D line plot for moving averages you can even rotate this 3D plot and if you want you can also create 3D bubble chart for strong day this is so powerful absolutely amazing and absolutely beautiful see this is for Google Apple and Microsoft and.
If you are curious enough you can also create 3D surface plot of stock price you can even rotate this also and this is the Apple price moment with annotations and apple with Ballinger bands and RSI and a correlation of stock returns you can see here it is really amazing and it becomes so easy Once you learn the pandas thoroughly so we will take time and you let me know if you want me to give enough time to Pand or not or if you just want me to just go through it and move to the next topic you can let me know in the comments so as I have mentioned already that any algo Trader or a Quant Trader gives 70% of their time in data analysis so you have to be proficient in data analysis right so we will give enough time to this Panda’s library and I will make sure that you learn each and everything and I will also make sure that you understand few of the secret tricks and tips that only few of the expert algo Traders knows and one more.
Thing that after watching this video please you decide that what you want you want me to quickly cover this topic or you want a deep understanding of this pandas right so with that being said let’s get started pandas in algorithmic trading pandas is a powerful data manipulation and Analysis library for python it provides data structures like series and data frame which are essential for handling and analyzing time series data such as as stock prices in algorithm trading so as we have previously learned that in numpy we have single major data structure that is nump and similarly in pandas we have two major data structures that is series and data frame and that is absolutely simple and you have seen those data structures since your childhood let me show you how Okay so as this line says it provides data structures like series and data frame so what is a series and what is data frame so simply.
If you have seen in the AEL Ms XEL we have a table right and inside that we have few columns and few rows right let’s say the First Column is serial number second is let’s say price of stock and third let’s say symbol of the stock and it could be the date right now this is simply a data frame that’s it a table means a data frame table is equals to a data frame as simple as that now this data frame consists of four columns and five rows right so any single column or a single row is known as the series that’s it that is the a basic a definition of Series so so let’s say if you take this row out of this table it will become a series simple similarly.
If you take out this row out of this data frame it will become another series means simply a single row or a single column is known as the series hopefully it’s clear if you have any doubt please get it cleared in the comments and now let’s understand some key features in pandas the first feature is data structure so it provides a that is 1D and data frame that is 2D for structured data right we have seen that this is the the whole table is 2D and a single row or column is 1D correct then time series analysis inbuilt support for handling and manipulating time series data because in trading we have the historical prices or even the live prices and all those prices are based on the time so it has a inbuilt support for time series analysis and that is really very very important so we will give enough time in understanding time series analysis right thenÂ
We have the data cleaning guys please understand this important concept that let’s say you have downloaded a data of last 10 years of any stock let’s say apple Bitcoin anything right so in that data of 10 years if you have any value that is not desired then it might break your whole code or it might alter all the results so you might get completely different results from what is your expectations so it becomes extremely important to clean your data so it has tools for handling missing data filtering and aggregating data right the fourth point is data alignment so it is so the pandas can automatically Align data based on labels making operations intuitive the fifth one is integration so it can easily integrate with other libraries like numpy mat blti and Skyy so.
We will learn few more libraries after the pendas and you will see that how it can easily integrate with those libraries then the last is IO tools so functions for reading and writing data from various formats such as CSV Excel SQL Etc means we can analyze or we can manipulate any kind of format like CSV Excel SQL and all those so now when we discussing SQL so let me know if you want a session on SQL because in algo trading you will need to have a good knowledge of SQL also right so let me know in comments that if you want a session on SQL or not now before moving further let’s compare numai and pandas right so as we understood that in pandas we have two major data structures that is series that is 1D and data frame that is 2D similarly in numaiÂ
We have an dimensional array that is andd array right again pendas is majorly used for data manipulation and Analysis whereas numai is used for numerical computation and operations so we will be using numpy also in the panda because in trading the whole game is about the time right because in trading the whole game is about the time so we have to make sure that we are time efficient pandas have excellent support for time series and that is required in algo trading and but numai has basic support again pandas have built-in functions for handling missing G right where is numai require custom handling pandas can automatically align the values based on the labels but in numpy.
We do not have the alignment because it is purely positional next pandas is slower due to additional features as we know that and as we have discussed previously that the more features we have it will become more slower that’s why ND have only one thing and that is only the values and because of that it is very fast and also it is based on the C right again panda have robust Group by and aggregation methods so group bu is also another object of pandas and we will learn that also thoroughly because it is also very important in Alor so we learn that also whereas numai has limited grouping capabilities again pandas integrates well with other data tools whereas numai serves as the computational backbone and we can easily integrate the pandas with mat PLO lip whereas numai requires external floating libraries again pandas have higher memory uses due to metadata right whereas numai is more memory efficient for row numerical data so that was the basic comparison and you should be having a good understanding of both the numai and pendas for efficient algo trading operation now let’s quickly dive into the interesting and let’s understand the pandas with live coding so the first thing is how to import pandas simply we can write import pandas as PD that’s it okay but as we know that we’ll be using numai also along with the pandas so it’s better always to import numai along with the pandas right so.
We just write import numai as NP correct so when you hit shift enter if it doesn’t throw any error that means you have written anything right and both of your libraries have been important now let’s create a series with list you know that a series is just a single row or a single column of any table right so series from list so what we will do we will create a series so let’s say list of string St Str correct then I can give as strings so let’s say the name of stocks so Apple then Microsoft then let’s see Bitcoin and ethereum correct now you have have to write PD dot this is a series class and inside that you just have to pass your list so list St Str correct and that’s it you have your Series so when you hit shift enter you will get the series but you might be wondering that I have just given these strings but how come this numbers are here 0 1 2 3 and that is the default index of any series or any data frame so whenever you create anything any series or any data frame without explicitly giving the index then this pandas will assign that index to your series okay now let’s move and create a series with integers so what I’ll do I’ll create a list of integers and let’s say the price so 100 1001 1002 1003 and I can give it the name like list integers correct and similarly you have to write PD series is and you have to pass the list so list integer and again when you hit shift enter you will see that you have your integer series with the default index given by the pandas now there are three things to observe the first is the index that is the default index by pandas second is the values these are known as the value right the value of the series and the third is the data type so here it was object and here it is integer 64 so you can say that in pendas a string is always considered as the object and the numbers can be
Watch this Day 23 video tutorial
Day 23: Pandas Series
7/20