What Are Time Series - Applied Time Series Analysis in Python and TensorFlow

👉Get the full course at 87% off: https://www.udemy.com/course/applied-time-series-analysis-in-python/?couponCode=TSPYTHON2021 Email me for a coupon if the one above expired: peixmarco@gmail.com ---------------------------------------------------------------- A time series is simply a set of data points ordered in time. Therefore, time is the independent variable. We can divide a time series into 4 different components. We have the level, which is the average value of the time series. Then, we have trend, which is the process that makes the values increase or decrease over time. Seasonality is a repeated cycle over time. Finally, we have noise, which adds randomness to the series. We will see how each component will guide us in our process of analysis and forecasting. Our objective in time series analysis is often to predict the future, but we might be interested in understanding different components of the time series, such as seasonality, or if there is autoregression. Of course, we will dive into those topics in depth later on. For now, let’s look at some examples of time series. Here, we see a simulated random walk, meaning that your time series is completely random! There is no real reason as to why it goes up or down. Here is another example of time series, where we have both an autoregressive and moving average processes in play. In time, we will go over what that means, and how to simulate those processes. Finally, here is a real dataset, which is the quarterly earnings per share of the company Johnson&Johnson. Notice, the trend here, since it goes upward. Also, we notice some seasonality, as the values go up and down, in a cyclical fashion. Those are all important elements that we will learn to identify and how that will impact our analysis.