Forecasting types of energy consumption
1. Forecasting types of energy consumption
In this scenario, I am a data analyst for an energy company, and I want to predict secular trends around renewable energy consumption to drive our internal production planning. Here, I have a dataset that shows the total energy produced each day for the two main types of energy: fossil-fuel based and renewable. Using the built-in capabilities within Power BI, I should be able to create a forecast that works for my purposes. To start, I’m going to create line charts to plot renewable energy and fossil fuel consumption, respectively. I will select the line chart visual from the visuals tray. Then, I will drag our date field to the X axis, and the renewable energy consumption total to our Y axis. Ill repeat these steps, instead replacing the fossil fuel consumption field as our Y axis for a second line chart. Within each of these charts, I will set the axis to be on the same scale, so we can have a better side by side comparison. After setting each axis to have the same scale, we can see that renewable energy has had a larger portion of total energy, and fossil fuel has been declining in recent months. Now, to create a forecast on these charts, we select each chart, go to the analytics pane, and turn on the forecast option. Within this pane, we have several options to fine-tune our forecasts.We can select the units of our forecast, ranging from date entities all the way from years to seconds. We will keep points as our option, as that reflects what we have in the data. From there, we can designate how far forward to forecast, and how much of our data to ignore. Ill set our forecast to ninety days in the future, and will keep all data in our forecast. Next, if we know any information about the seasonality of our data, we can incorporate that. We won’t need that for our analysis, but it’s good to know it’s an option. Finally, we have a confidence interval to plot, which will show a range of values that have a given likelihood to be accurate. In my case, I would rather try to hone in on a more narrow range of potential consumption values so that I will drop our confidence interval to 80%. Comparing our forecasts, we can see that our initially observed trends are likely to continue. Renewable energy consumption will rise, and fossil fuel consumption will continue to fall. This was a short example of how to quickly get started using the built-in forecasting capabilities within Power BI. Now, let's use these techniques on some financial stock data!2. Let's practice!
Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.