Time series data is used to estimates about the future of our data using forecasting option. Using forecast option we can set goals and estimates for future.
Forecasting uses the Exponential Smoothing Models. With exponential smoothing, recent observations are given relatively more weight than older observations.
In today’s blog I will explain how we can use forecasting in Power BI.
Forecasting option is available in Analytics pane for Line charts. Currently Forecasting is available only for Power BI desktop not for Power BI Services.
Forecast is exist only for single line, we cannot use forecast option for multiple line charts.
I have loaded Sample Super store data set in Power BI and created a line chart to analyse Profit (Y axis) with respect to order date (time series- X axis)
Use hierarchy and display data values for month wise.
Power BI visualization:
Select Analytics pane (keep this visualization selected)
Click on Add option for Forecast:
Forecast line is drawn and we can see a grey area with respect to forecast line. With forecast values in tooltip , we have upper bound and lower bound values too.
For forecasting we have few different options, let’s have a look for those:
Forecast Length: By default, value will be 10 points in future. We can change this range according to our requirement that for which period we want to find out forecasting, in this example I am going to forecast for 12 months
Ignore last: How many months we want to ignore. We can keep this blank or can give any values. e.g. if you do not want to use values from last one month to evaluate forecasting, we can set 1-month value in Ignore last field.
Confidence interval: We can change confidence interval; confidence interval is that if we run this test on different set of data we are going to get 95 % of same results as forecasted.
Seasonality: Expedition smoothing is looking into seasonality and asking for points, we need to look one cycle of data values present in our visualization.
In my example in one cycle of data is 12 months and we are forecasting in months so I will put 12 points in seasonality.
Let say if I have data for every single day and our one cycle is of 12 months, in that case we should put 365 points for seasonality
Now let’s change:
Forecast length to 12 months
Ignore last: blank
Confidence interval: 95%
Seasonality: 12 and Apply
Now you can see Forecast line has taken shape in place of a start line in very similar trend that look very similar to trend that existed in last.
We can do more formatting for our forecasting:
If we do not want to display lower and upper bound area band we can set Confidence band style to None:
Now let say we have data points for daily.
As now we have data values for every day and there are 365 days in one year i.e. one cycle in our data set , set 365 points in seasonality.
Reach out to us if you are looking for Corporate Training to Build The Next Generation Analytical Workforce with an in-depth understanding of Exploratory Data Analysis , Data Visualisation, Data Analytics , AI First , Machine Learning & Deep Learning Training & Consulting helping them to take Data Informed Decision at each stage of the business.
We understand that At the present times , the Entire Industry is in a Tranformation stage with the Softwares being rebuilt with Artificial Intelligence Capabilities .
We need SMART WORKFORCE for the SMART SOFTWARES to reap the maximum return .
Whatsapp at +91-9953805788 or email at - firstname.lastname@example.org if you would like to know more .