From the course: Time Series Modeling in Excel, R, and Power BI
What you should know
From the course: Time Series Modeling in Excel, R, and Power BI
What you should know
- [Instructor] Time series modeling is an area of data science that uses regression models to make predictions about future trends. As time passes and we get more data points, we can then incorporate this additional data into our model as feedback loops. Models like this are part of the machine learning space and the greater AI landscape. To do this course, you should have an Excel license. You'll also want to enable the data analysis add-in in the options menu. If you want to try out the R and Power BI models in this course, you'll also want to download them online. The good news is that both their desktop versions are free. You can download R from its website. We'll also use R in tandem with RStudio as the IDE, or integrated development environment, to test that our R code runs. You'll download RStudio from the Posit website. If you have multiple versions of R on your computer, you can set the version you're running in RStudio through the go-able options selection in the tools dropdown menu. To download packages, you can install them through the R Query or you can install them directly in Rstudio, using the function install.packages, followed by the name of the package. When you run this single line of code, you'll see the package installed in the console below. If you'd like to try out Time Series Modeling in Power BI, you can download Power BI Desktop for free online. I'm using the December 2022 version of Power BI Desktop in this course. You'll then need to enable R directly in Power BI Desktop through the R scripting settings menu and the options menu within the application and point it to the version of R you're using on your computer. When you run R in Power BI Desktop, it runs off the version of R on your computer, but if you publish it into the cloud accounts in the Power BI service, these pro and premium accounts support the R packages on the Power BI documentation page we see here. All the packages we'll use in this course are supported in the Power BI Service. In our project, we're going to make some assumptions about doing time series models. Typically, testing algorithms like this involves splitting the data into training and testing sets, but in our project, we're going to simplify the modeling process by eliminating this step. We're also not going to remove anomalies or outliers from the data. Focusing on the key parts of time series modeling will give us the chance to really play around with these data models.
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.