This is the start of an educational series of articles on machine learning and its application in securities trading, more specially cryptocurrency trading. We will not be getting into the calculations although maybe in a future series we could do so. This is to give you a broad base understanding of the techniques in the fields of machine learning. We would encourage you to use this as a steppingstone to further learning in the subject. We will start with regression.

**What is regression?**

Regression analysis is a statistical technique used to examine the relationship between one or more independent variables and a dependent variable. The goal of regression analysis is to understand how changes in the independent variables are related to changes in the dependent variable.

Regression analysis is used for a wide range of purposes in various fields, such as finance, economics, social sciences, and engineering. It is particularly useful for predicting future outcomes based on past observations or data. Regression analysis can be used to find trends and patterns in data, to estimate the strength and direction of relationships between variables, and to make predictions about future events or trends.

In regression analysis, there are two types of variables. These are ‘independent’ and ‘dependent’ variables. The independent variable is a variable that is manipulated or changed to observe its effect on the dependent variable, while the dependent variable is the variable that is being observed or measured to determine the effect of changes in the independent variable.

Graph: relationship between dependent and independent variables

**A simple example…**

To provide a simplistic example, suppose we are interested in predicting the salaries of employees based on their years of experience. In this case, years of experience would be the independent variable, as it is being manipulated or changed to observe its effect on salary, which would be the dependent variable.

In this example, we might collect data on the years of experience and salaries of a sample of employees. We could then use regression analysis to model the relationship between these two variables. The regression model would estimate how much of the variation in salaries is explained by years of experience and could be used to predict the salary of an employee based on their years of experience.

Another example might be to predict the price of a particular cryptocurrency based on various independent variables such as trading volume (the number of coins bought and sold on an exchange), market capitalization (the value of all the coins that can be traded), and sentiment analysis (the views of individuals at any one time on a specific cryptocurrency, cryptocurrency market or in general). In this case, the independent variables would be manipulated to observe their effect on the dependent variable, which is the price of the cryptocurrency.

Regression analysis can be used in two main types: linear and non-linear. Linear regression analysis is used when there is a linear relationship between the independent variables and the dependent variable. Non-linear regression analysis is used when there is a non-linear relationship between the independent variables and the dependent variable.

**Applications of regression analysis.**

*Prediction:* Regression analysis can be used to predict future trends and events. For example, it can be used to predict the stock market's performance based on economic indicators, or to predict the demand for a product based on historical sales data.

*Estimation:* Regression analysis can be used to estimate the value of a dependent variable based on the value of one or more independent variables. In the above example we could estimate salary for someone with 6 years’ experience to be around $40k.

Graph: using the regression line to estimate salary based on experience.

*Hypothesis testing:* Regression analysis can be used to test hypotheses about the relationship between variables. For example, it can be used to test whether there is a significant relationship between a person's income and their level of education. A regression analysis has a number of statistics associated with it of which significance of the relationship between variables can be determined.

Overall, regression analysis is a powerful tool for understanding and analysing the relationships between variables, making predictions, and testing hypotheses. It can be used to gain insights into a wide range of phenomena and is an essential tool for researchers and analysts in many fields.

What we have so far described is linear regression is a statistical technique used to model the relationship between two variables. It is often used to predict the value of one variable (the dependent variable) based on the value of another variable (the independent variable). Linear regression assumes that there is a linear relationship between the two variables, meaning that as one variable increases or decreases, the other variable changes proportionally.

**The next level – multiple regression.**

Multiple regression is like linear regression but instead of having one independent variable you can have more than one. It is used to model the relationship between multiple independent variables and a dependent variable and can help to identify which independent variables are most strongly associated with the dependent variable.

In cryptocurrency trading, linear and multiple regression can be used to identify patterns and relationships between variables that might be predictive of future price movements. For example, a trader might use linear regression to model the relationship between the price of Bitcoin and the trading volume of Bitcoin over time. If the analysis reveals a strong positive correlation between these variables, it might suggest that higher trading volumes are associated with higher prices, and vice versa.

Similarly, a trader might use multiple regression to model the relationship between the price of Bitcoin and multiple independent variables, such as trading volume, mining difficulty, and the sentiment of social media posts about Bitcoin. By analysing these variables together, the trader may be able to identify which variables are most strongly associated with changes in Bitcoin's price and use this information to make more informed trading decisions.

It's important to note that regression analysis is not a fool proof method of predicting future price movements, and that there is always a risk involved in cryptocurrency trading. Additionally, regression analysis is based on historical data, which on its own may not accurately reflect future market conditions but can be used a part of a larger model. As with any trading strategy, it's important to do your research and understand the risks and techniques fully before making any investments. At Neomony we have years of experience dealing with all manner of statistical tools including regression and applying them to strategies in a robust way.

If there is interest in this series of blogs, and it is requested we would be happy to increase the detail to help you understand machine learning concepts like regression and how we further apply these concepts at Neomony.

Find out more about how Neomony can help you invest in the crypto currency markets !

## Comments