Hi. I'm Gideon Ozik. Hi. I'm Sean McOwen. I'm a research associate at EDHEC-Risk Institute, Co-founder and Managing Partner of MKT Media Stats, a research and analytics company that produce alternative data sets for financial market applications. I'm currently quantitative analyst at 74 Capital Management, as well as a candidate for a Master's of Analytics to Georgia Tech. I've been working with Gideon for about a year and a half on some educational resources around data science and quant finance, and we're really excited to share with you. Welcome to the course. People have been using data to involve their investment decisions for many decades. Traditional data sets such as price, return, volume, volatility, are just a few examples that people use to involve their decision-making process or decision-making model. In this class, we'll focus on alternative data. We'll actually going to cover four types of alternative data. In the first week, we'll cover consumption. Consumption is the type of data that allow us to understand where people are? What they shop? What brands they are interested in? From that, to derive insights to inform our decision-making process. In the second week, we'll cover the open web. This is information that will extract directly from the Internet to get an idea of how the economy's doing to develop some insights about different firms and different sectors. In the third week, we'll cover corporate and regulatory disclosures. For example, we look at financial statements of companies and filing that they do with the regulators in order to extract information about their state, about the connection with other firms, and so on and so forth. In the last week, the fourth week, we'll focus on the media. By media, we refer to both press, digital media, and also social networks. Now, the structure for this course is that each week is flipped to three sections. First, we go through the theoretical. Now this is going to be your equations, the actual stuff that you're eventually going to implement. Next, we go into the lab. This lab is where you'll take a real data set and you'll be programming in Python, and you'll see how we go through step by step into taking a large unclean data set into actual insights. Finally, this third piece of that puzzle is going to be the application section where we look at the research and what people have been doing for the last 20 years in the field of alternative data and the ways that they've used it to predict returns, risk, and much more. Should you take this MOOC? Well, if you're looking to enhance your analytics skillset as they apply to the financial markets, you should definitely consider taking this MOOC. If your objective is to gain hands-on experience in addition to theoretical concepts, you should definitely be here. If you're excited about cutting edge technology and research as they apply to big data and alternative data sets, this is definitely a place for you. Together, if you took this MOOC, you will enjoy high career prospects as a data scientist in financial markets. What should you expect to know at the end of this MOOC? First, you'll understand what is geolocation data, how to process it, and how to transfer it into a form that will allow us and you to extract valuable information from it? You'll understand how to parse text and transfer the text into a quantitative format. Then you understand how to measure a tone of a text. Therefore, by using sentiment analysis technique to quantify the information delivered by textual content. Finally, you understand how to build networks and how to derive insights through the analysis of networks. What skills can you expect to learn from this course? Number 1, you've got to work with large data sets and take it and reduce bias, look for where you can compress some of the issues and then move on to actual insights. They're also going to learn web scripting tools. These are the tools that let you go to the Internet and take down whatever information you need automatically, it's really helpful when you need to get a large amount of data. We're also going to work with creative problem-solving for real-world data issues. A lot of times, data in the real world has a lot of noise, a lot of different things that you do to actually work out that our errors, and we're going to work through some creative solutions to dealing with it. We're also going to go through advanced visualization. This is going to be some geographic mapping and other advanced methods you might not have seen yet. Finally, we'll talk about how you actually get the economic intuition from the data once you have it. So these are the core skills you will be working on in this MOOC. There are four primary data set types. Now, first is consumption data. This is an Uber data set that we've gotten from Kaggle. This Uber data set will go through all the different locations of Uber rides, and try and figure out insights from that. We also have the open web. Now, this is Wikipedia. We're taking the different textual data from Wikipedia and turning it into actionable insights. There is also corporate disclosures, and we're going to go through how you get data from Edgar and parse it into textual data form. Finally, we're taking another data set of Twitter data from Kaggle, and we're going to be using this for the networks. We hope to see you in our class. We think you'll learn a lot.
