Variable selection

In the dynamic world of forecasting, choosing the right variables, also called leading indicators or features, can make all the difference between accurate and unreliable forecasts. Our advanced variable selection tool empowers you to identify the most relevant variables, optimizing your forecasting models for accuracy, efficiency, and reliability.

Correlation vs. advanced methods

Key features in Indicio

• Automated Feature Importance Analysis
  ‍Harness the power of machine learning to automatically rank variables based on their impact on forecast accuracy. Save time by allowing our system to highlight the most critical variables, ensuring you focus on the ones that truly drive results.
• Manual override
  While automation is powerful, you know your data best. Easily override automated selections by manually selecting or deselecting variables. This gives you full control over the forecasting process and enables customization for specific business needs.
• Cross-Validation
  Test different sets of variables against one another using built-in cross-validation. This ensures that the selected variables performs well on data that the model hasn't seen yet (out-of-sample), improving the model’s predictive performance.
• Time Series Feature Engineering
  Automatically generate lag features, moving averages, and seasonal factors relevant to time series forecasting. Easily adjust these features to match your data’s granularity and forecasting horizon, improving long-term accuracy.

Webinar - Identifying leading indicators

How it works

• Create a Forecast
  Begin by setting up your forecast. Give it a unique name and define the forecast horizon, specifying how far into the future you want to predict—whether it’s days, months, or even years.
• Import Data
  Seamlessly bring your data into the platform by uploading files, fetching data from our integrations with 3rd-party data vendors, or connecting directly to your database. Indicio supports a wide range of formats to ensure flexibility.
• Analyze Variable Importance
  Let Indicio automatically analyze and rank the importance of each variable. Our system evaluates which features are most likely to impact forecast accuracy, helping you focus on the key drivers of your model.
• Build Models
  Choose from an extensive library of both statistical and machine learning models to build your forecast. Whether you prefer time-tested approaches like ARIMA or cutting-edge neural networks, Indicio has you covered.
• Evaluate Models
  Use cross-validation to evaluate your models and get out-of-sample accuracy. Compare models based on the metrics most relevant to your business needs—RMSE, MAPE, MASE, or Hit-ratio—and choose the best performer for deployment.
• Export Forecast
  Once your forecast is finalized, easily export it to your IT environment. Share the results with your team to help your organization make informed, data-driven decisions with confidence.

Explore more features

Build models

Building forecasting models doesn’t have to be complex!
In Indicio, it's just a few clicks

Scenario Analysis

Simulate the effect on your business, if inflation increases or a slowdown in the economy

Explainable Forecasting

Translate complex forecasting models into clear, actionable insights