What are Trading Robots? (And why should I build them?) (Part 1)

In this session, we're going to talk about three things. What is algorithmic trading that will define our version of algorithmic trading, that we are going to talk about our tools and do a little bit of a demonstration at the end to show you what this court can offer, essentially, algorithmic trading is trading done by the computers. The computer will be the one to analyze the market, after which the computer will be the one that initiate, buy and sell signals, and they will be the one that manage operations and treats. There are two main types of algorithmic trading. First is high frequency type, second low frequency type, high frequency trading can be characterized as treeing where your advantage is in the speed of your connection, let's say the same, go there. But if it's still at one air change and another change, a different process, if your computer system and connection is fast enough, you can buy the cheaper one and sell the more expensive one instantly. That is called arbitrage. In high frequency trading, the amount of trade is usually minimal, but they make up for it in bulk. Millions of trades goes on every month in a high frequency trading firm, and they could trade thousands of times in less than a second. Low frequency algorithmic trading is characterized as trading with advantages in a trading model in low frequency trading. The speed of your computer system is not that important. When we do our algorithmic trading robots, there could be many different types of logic that we use, say we could use macroeconomic news. These are economic data releases like the non-farm payroll or the FOMC policy company fundamentals, revenues, earnings, etc. or we could use statistical methods like correlation, Russia and integration, etc.. Technical indicators, technical indicators are algorithms that process the data giving us output that could allow us to view the market in another way that offers us in each market microstructure can be defined as the brand of finance that is concerned with the details of how to assess change occurs in the market. For example, we could be looking at a structure and a design of the asset has changed. Who are the middlemen, who are the dealers and how is transport, etc.? How is price form and discover? Is it Ossian? Is it negotiated? Well, the transaction and processing cost of the trips and what is the transparency of the market information? By understanding the market microstructure, we can develop trading models to take advantage of it. It could be looking at the order on the floor, a bit crushed, etc.. There are too many ways to go about developing algorithmic trading robots, also known as E.S. or aspect of ISIS. So for this course, we will be focusing on low frequency mathematical models using market ideology. So let's establish what this means, market policy. This means that our ideas must be fundamentally sound from both a market and an economy point of view. Therefore, we can't say that just because a technical indicator may provide somehow predictive value. We're going to use it blindly when we develop robots in massive market. PRUDEN It must make sense from a market point of view. We must know why our robot works or why it doesn't. Mathematical models after we have an idea there is market pruden we test in optimize it using a proper statistical framework. This will have us develop robots that have positive expectancy in the long run and has proper risk management low frequency, the trading frequency of our models are less than once a minute. Our strategy does not depend on the speed or the computing capacity of our hardware, software or train systems. Therefore, we are able to execute our models and our robots using basic software and hardware, the ability to attract a kind of fun. To design vectors and execute our robots, we need to know three trenches, first is a tooth. What platform are we using? How are we going to court our robots? What performance analysis software we use to analyze the performance of our trading in the long run by software that we need to execute our robot? The second data, how to manage data, the sources where we get our data from and how to clean that data, etc.. The most important part will be our design theory. How do we come up with market Pruden ideas? How do we validate it, how we buy to optimize our work, and how the statistics and finance get incorporated in our ideas. And after we lost a product, how do we maintain and improve it as it goes along? These are the questions that are COSO answer. We're going to use Metacritic for the coding languages, MKR for and we are going to use Excel.