Trend prediction with AI

The ONLYOFFICE Spreadsheet Editor allows you to use AI assistants to predict future trends based on your historical data. By combining AI capabilities with the built-in forecasting functions, you can generate predictions, identify seasonal patterns, and make data-driven projections for business planning, financial forecasting, and other analytical tasks.

By default, AI is not connected to editors for security reasons. To use AI for trend prediction, you need to install and configure the AI plugin first. Refer to the AI assistants: setup and configuration guide for detailed instructions.

Accessing AI features

To open the AI prompt in the Spreadsheet Editor, use one of the following methods:

Go to the AI tab on the top toolbar and select Chatbot.
Right-click the selected data and choose AI → Chatbot from the context menu.
Press Ctrl+/ (or ⌘+/ on macOS).

Using AI for trend prediction

Analyzing historical data for patterns

Before making predictions, use AI to understand the patterns in your historical data:

Select the cell range containing your historical data (e.g., dates and corresponding values).
Open the AI prompt.
Ask AI to analyze the data for prediction purposes. For example:
- "Analyze this historical data and identify any trends or patterns"
- "Is this data suitable for linear forecasting or does it show seasonality?"
- "What patterns should I consider before making predictions?"
- "Describe the growth rate and any cyclical patterns in this data"

Getting AI recommendations for forecasting methods

AI can help you choose the right forecasting approach:

Select your historical data.
Open the AI prompt.
Ask for forecasting method recommendations:
- "Which forecasting function should I use for this data?"
- "Should I use linear regression or exponential smoothing for this prediction?"
- "What's the best way to forecast this data considering its seasonal pattern?"

Generating prediction insights with AI

Use AI to interpret your forecasting results and provide context:

After calculating predictions using forecasting functions, select both your historical data and predicted values.
Open the AI prompt.
Ask AI to interpret the results:
- "Interpret these forecast results and explain what they mean for planning"
- "What is the confidence level of these predictions?"
- "What factors could cause actual results to differ from these forecasts?"
- "Summarize these predictions in business terms"

Built-in forecasting functions

The Spreadsheet Editor provides several statistical functions for trend prediction. Use AI to help you select and configure the right function for your data.

FORECAST.LINEAR — use for data with a consistent linear trend (steady increase or decrease without seasonal patterns).
FORECAST.ETS — use for data with both trend and seasonal patterns. This function uses exponential smoothing (ETS algorithm).
FORECAST.ETS.SEASONALITY — use to detect the seasonal pattern length in your data before applying forecasts.
FORECAST.ETS.CONFINT — use to calculate the confidence interval for your predictions, giving you a range of likely values.
TREND — use to calculate multiple predictions along a linear trend line at once (array formula).

Visualizing predictions with trendlines

You can visualize your predictions by adding trendlines to charts:

Create a chart from your historical data (go to Insert tab → Chart).
Select the chart and access the chart settings.
Add a trendline to visualize the trend direction.
Enable Display equation on chart to see the trend formula.

Trendlines help you visually validate that your forecasting function results align with the overall data pattern.

For more information on creating charts, refer to Insert charts.

Limitations and considerations

When using AI and forecasting functions for trend prediction, keep the following in mind:

Historical patterns may not continue: Forecasts assume past patterns will persist. Major market changes, new competitors, or economic shifts are not captured.
AI interpretation vs. calculation: AI provides insights and recommendations but does not perform the mathematical forecasting. Use the built-in functions for precise calculations.
Data quality matters: Missing data, errors, or inconsistent recording periods can significantly affect prediction accuracy.
Confidence decreases over time: The further into the future you predict, the less reliable the forecast becomes.
External factors: Neither AI nor mathematical functions can account for unknown future events. Always apply business judgment.

Warning: Predictions are processed using built-in statistical functions and interpreted by an AI service of your choosing. ONLYOFFICE assumes no responsibility for business decisions made based on forecasted values. Always validate predictions against business knowledge and consider multiple scenarios.