Author(s: Gift Ojeabulu) Learn how to perform statistical analysis using python and create insights like a professional, using FIFA 21 for a case study. Guglielmo Basile on Unsplash. A brief story about why I chose to write this article. This is a man who wanted to learn data science. The internet provided him with many resources, including YouTube, Coursera and Udemy. He worked on a small, clean dataset through tutorials. This excitement led him to learn a lot more about data science and solve real-world issues. If you’re familiar with football, you will know that it is not an easy game for Sheffield. Real-life data was messy and it caused him to feel like a fraud. He learned that he needed to deal with missing values and ask intelligent questions. Later, he realized it wasn’t just about performing analysis. EDA wasn’t well explained, particularly at beginner level. This could be because the tutorials don’t have a pattern or other reasons. He learned it through many resources and practice, solving real-world issues, and simplified exploratory data analysis. These are some tips and code examples using the FIFA 2021 data set. My talents are not special. I am only passionately curious — Albert Einstein. Passionately curious means that we are constantly learning and becoming more curious about data. This is why, when playing FIFA 2021,, I wanted to know the answers to specific questions such as: Which players are the fastest in FIFA 2021?? What are the highest-paid players? Which players are the tallest? Which players are the strongest? What are the top long pass players? What are the most skilled players who can make short passes? What are the greatest defenders? This tutorial will teach you all about exploratory data analysis. This article will explain: What is exploratory data analysis? What is Exploratory Data Analysis? Checking missing values and renaming columns What is Exploratory data analysis? Exploratory Data Analysis (EDA), is a method of analyzing data to determine their key characteristics. It uses visual methods. Although a statistical model is possible, EDA is primarily used to see what data sets can reveal beyond hypothesis testing or formal modeling. Source: Wikipedia. Exploratory Data Analysis is an attitude. It is the willingness to search for things that aren’t there and those we think might be. John Tukey Explorer Data Analysis is, from my point of view, a method of inspecting, inspecting, and turning data into useful information to support rational decision-making through statistical techniques or data visualization. What is Exploratory Data Analysis? EDA’s main function is to examine data without making assumptions. EDA can identify errors and help to understand the patterns in the data. It also helps detect anomalous events or outliers, and find interesting relationships among variables. Source: Ibm.com. I love Exploratory Data Analy because it allows me to ask smart data questions and better analyze and visualize data datasets in order to find patterns. It also provides valuable insight that can be used for data science’s next steps. This allows me to think critically and beyond my intuition. It also helps with problem-solving and analysis. Understanding the Data Sets Before you start any data science project it is important to understand and prepare for its data. If you don’t have any knowledge about the data, it is impossible to perform an effective analysis. Therefore, I recommend that you have a good understanding of the data. To help you understand your data, I have created a variable description that will allow you to dive into exploratory data analysis. Let’s now get to the practical part of this article, once we have a better understanding of exploratory data analysis. The FIFA 2021 data set, obtained from Kaggle, will be used. You can find the GitHub repo that describes the dataset as well as the notebook. You will receive a folder containing the data and the notebook. To better understand the article, I recommend opening the notebook. The notebook should be simple and well-commented, organized, and detailed. Link to Detailed variable description here https://medium.com/media/7c35b59c8a602a68541141cb220eaa1f/href We importPandas,matplotlib numpy ,seabornfor basic data manipulation, numerical computation and visualization. The filterwarnings module is used to silence any warnings. Word Cloud can be used to represent text data at different sizes, from most frequently mentioned words to least. To visualize the text data, Word Cloud is used to import it. Counter holds the word-frequency and counter the element count. Finally, the CSV file is read and viewed with the.sample function. This will give us a random view of the dataset. We can see that FIFA contains 3 types of features: Date, Number, and Categorical. Let’s now do some data cleansing and feature creation. These techniques are intended to accomplish the tasks described in this article. You can choose the method that suits your needs or preference. Renaming Columns & Checking for Missing Values #Viewing the columns available in our Datasetdata_fifa.columns #Checking for columns with missing valuesdata_fifa.columns[data_fifa.isnull.any] We use the .columns from Pandas to show the columns in our dataset, we could deduce that we are dealing with a dataset with a lot of columns. To check for missing columns in the dataset, we use the.columns [data_fifa.Isnull.any]. We found three missing columns. Common errors in real-life data include spelling and punctuation mistakes, inaccurate data with fields, duplicate data or data that is not current or complete. This can adversely affect the data and can impact our analysis. It is important to address these problems in our dataset. https://medium.com/media/75c3e39ee5a12a36af07260721e98bad/href In this part, we are renaming our column name to a shorter and well descriptive name, a good practice for Data scientists, where we adopt the Camel case Naming convention. DateTimesupplies Classes are available to help you work with dates and times. The classes offer several functions that can be used to handle dates, time and intervals. Python has Date object and DateTime object. This means that you can manipulate these objects, not strings or timestamps, when manipulating them. https://medium.com/media/16470f410aa21701796113fc7830aa9c/href We are simply transforming the object column to the object column and then we are using the DateTime properties to transform the
