Random Walk Model - Applied Time Series Analysis in Python and TensorFlow

👉 Get the course at 87% off: https://www.udemy.com/course/applied-time-series-analysis-in-python/?couponCode=TSPYTHON2021 📚 Link to the notebook: https://github.com/marcopeix/AppliedTimeSeriesAnalysisWithPython/blob/main/HOTSAP_random_walk.ipynb Email me for a coupon if the one above expired: [email protected] ----------------------------------- Let’s now introduce the random walk model. The random walk model states that the location at time t is the sum of the previous location and some random noise. In this case, we assume that the noise is normally distributed, so it has a mean of 0 and a variance of 1. Of course, if we start the random walk at 0, then any point in time is the sum of the noise, expressed like this. Looking at the ACF plot for this time series, we see that the autocorrelation is very high at first, and slowly decreases. This is indicative of a trend. Now, is there a way to remove that trend? Of course! Since we know that random walk adds noise to the previous point, if we take difference between a point and a previous one, then we get only noise! If it seems unclear now, don’t worry, we will go through each step in Python, in the next lesson. So, by taking the difference and plotting the result, we get the following: noise. We see that there is no clear trend in the time series, and we it is purely random noise.