May 11, 2021
What’s new with FORECAST Squared?
The FORECAST² team is always looking to see what we can do to improve our users’ forecasting experience, and lately, we’ve been quite busy! We thought it was about time that we update everyone on some of the latest features that we added to give you the best of FORECAST²
Aside from the customizable dashboard, we introduced with this new release, let’s look at some of the recent updates, upgrades, and features that make forecasting an even more fun experience.
- An all-new Admin Portal
Unless you are an Admin you may not really see this feature but trust us when we say the Admin Users now really love it. The new “tile” view makes user and license management much more intuitive and easy to use.
- Self-Evaluating forecasting accuracy
There’s no need to save and compare forecasts with actuals in Excel anymore! Now our users can self-evaluate forecasts against the actuals in FORECAST2 itself. All you have to do is click on the compare icon and generate a report on accuracy. You can do this for a single forecast or a collection of multiple forecasts with just one click!
We even added a warning sign to make it easier to identify forecasts that are not performing well, so you can either re-train the model with the latest data or bring in other predictors to better the accuracy.
- Keep track of your modeling history
Our customers have been asking us to help them track their re-training and re-forecasting activities for a while now. And of course, we obliged! Now you can see the historical timeline of your re-training and re-forecasting activities in the viewing panels. You can also see that information when you compare your forecasts with the actuals. Now you know exactly when to re-train and re-forecast without any other additional information, note pads, or sticky notes!
- Expert Option for better control
This new feature allows demand planners and forecasters to forecast with a lot more precision & have higher control of their forecasts. We wanted to give some of our expert users with a higher degree of Data Science skill the opportunity to use their expertise a lot more
Some of the notable introductions to this option include selecting;
>> the desired classes of candidate algorithms: Expert users can choose from either Machine Learning or Time Series algorithms to get the best out of FORECAST2 for your forecast
>> different outlier adjustment methods
>> the best-fit criteria out of methods such as RMSE (Root Mean Square Error), MAD (Minimum Absolute Deviation), and MAPE (Minimum Absolute Percentage Error)
>> the best cross-validation method
- Collaboration forecasting
FORECAST² also allows you to share your forecasts with other relevant team members and adjust forecasts based on their feedback. You don’t have to spend time downloading forecasts before sharing them with your colleagues. You can share your forecasts directly from FORECAST² to as many members of your team as you like.
So what’s next?
The FORECAST² team is working on a fully revamped VIEW and VALIDATE forecast feature that is guaranteed to make forecasting withFORECAST² even more convenient.
Meanwhile, don’t forget to check out the webinar we hosted last month about the features of FORECAST² 3.3.