What is Tableau?
Tableau is an excellent data visualization and business intelligence tool used for reporting and analyzing vast volumes of data. It is an American company that started in 2003—in June 2019, Salesforce acquired Tableau. It helps users create different charts, graphs, maps, dashboards, and stories for visualizing and analyzing data, to help in making business decisions.
What is the purpose of forecasting?
Forecasting is a technique that uses historical data as inputs to make informed estimates that are predictive in determining the direction of future trends. Businesses utilize forecasting to determine how to allocate their budgets or plan for anticipated expenses for an upcoming period.
How to perform Forecasting in Tableau?
Dataset used in the given example is superstore.
Follow the steps below to open the worksheet:
Open the tableau tool and connect superstore.xls to it. Drag and drop sheets of the connected dataset.
Click on sheet1 to open the tableau worksheet
Drag the Order Date to the Columns
Need segment-wise sales so drag Segment and Sales to the Rows to find the line chart
For having region-wise sales, Now drag the Region to color under Marks as shown in the image below
These are four years of sales and to have both dimensions to predict next year's sales. Go to the Analytics tab and select the forecast under the Model category
Tableau will automatically add a line to your chart forecasting the sales for 2022. The light color around the line is the estimated range for sales.
We can predict for more years as well. Now click on Forecast -> Show Forecast to enable forecast for the visual.
You can set the Forecast Length in this menu and do the prediction for the next 3 years.
There are also options to adjust the Forecast Model as shown below and select Automatic.
Now we can see the range of the sales till 2023 by giving forecast length until 3 years.
Information about Forecast Options:
There are various selections related to forecasting that are available in the forecasting options that can be specified according to the requirements posed.
Forecast Length
The time frame for which the forecasting is to be done can be set using the forecast length option that can be set as automatic which takes up the next 5 quarters by default, the exact number of time periods, or the until time for the values to be calculated.
Source Data specific
Few features that are specific to the source data like the aggregate by an option that specifies the granularity of the time series. The default option is automatic which takes up the best granularity for the estimation. For forecasting done with integer dimension instead of Date, the aggregate by is always set to Periods. Also, you can set the last few frames in a time series that can be ignored to perform forecasting.
Forecast Model
The model to be produced is specified in the forecast model option. There are options like an automatic that takes up the best of the models, an automatic with no seasonality that takes up the best with no seasonal components, and a custom model where you can specify the trend and season
Hence these are simple steps to perform forecasting in tableau.