ElyxAI
formulas

How to Use FORECAST Function

Excel 2010Excel 2013Excel 2016Excel 2019Excel 365

Learn to use the FORECAST function to predict future values based on historical linear trends. This tutorial covers syntax, data preparation, and practical applications for sales forecasting, trend analysis, and budgeting. Master linear regression predictions to make data-driven decisions with confidence.

Why This Matters

FORECAST enables accurate predictive analytics for business planning, helping professionals anticipate trends and make informed decisions based on historical data patterns.

Prerequisites

  • Basic understanding of Excel functions and cell references
  • Historical dataset with at least two columns of numerical data
  • Familiarity with linear relationships in data

Step-by-Step Instructions

1

Prepare your data

Organize historical data in two adjacent columns: independent values (x) in one column and dependent values (y) in another. Ensure data is numerical and sorted chronologically.

2

Click the target cell

Select the cell where you want the forecast result to appear, typically below or adjacent to your data range.

3

Enter the FORECAST formula

Type =FORECAST(x, known_y's, known_x's) where x is the value you're predicting for, known_y's is your historical dependent data, and known_x's is your historical independent data.

4

Specify cell ranges

For example: =FORECAST(B15, B2:B14, A2:A14) predicts a value at position B15 using historical data from rows 2-14. Use absolute references ($) if copying the formula.

5

Press Enter and verify results

Press Enter to execute the formula and see the predicted value. Review the result for reasonableness and adjust data ranges if necessary.

Alternative Methods

FORECAST.LINEAR function

In Excel 2016 and later, use FORECAST.LINEAR as an updated alternative with identical syntax and improved naming clarity.

TREND function

Use TREND for array predictions across multiple values simultaneously, returning multiple forecasts in one operation.

Trendline in charts

Add a trendline to an XY scatter chart (Chart Design > Add Chart Element > Trendline) and display the equation for visual trend analysis.

Tips & Tricks

  • Ensure your known_x's and known_y's ranges are the same size for accurate calculations.
  • Use consistent data intervals; irregular time periods may reduce forecast accuracy.
  • Plot your data first to visually confirm a linear relationship before using FORECAST.
  • The x value doesn't have to fall within your historical range—FORECAST extrapolates beyond existing data.

Pro Tips

  • Combine FORECAST with confidence intervals using STDEV and T.INV to quantify prediction uncertainty.
  • Use FORECAST.ETS instead for seasonal or exponential smoothing patterns in your data.
  • Create sensitivity analysis by copying the formula horizontally with varying x values to model multiple scenarios.
  • Validate forecast accuracy against actual future data to refine your historical dataset and improve predictions.

Troubleshooting

Formula returns #DIV/0! error

This occurs when all x values are identical. Ensure your independent variable (x) has varying values to establish a trend line.

Forecast result seems unreasonable

Verify data quality and check for outliers that skew the linear regression. Consider removing anomalies or using a shorter recent data range.

Formula returns #N/A error

Check that known_x's and known_y's ranges are equal in size and contain only numeric values with no empty cells.

Copied formula gives incorrect results

Use absolute references ($ signs) around your data ranges when copying: =FORECAST(A2, $B$2:$B$10, $A$2:$A$10).

Related Excel Formulas

Frequently Asked Questions

Can FORECAST predict non-linear trends?
No, FORECAST uses linear regression and assumes a straight-line relationship. For curved or exponential patterns, use FORECAST.ETS or polynomial trendlines instead.
What's the difference between FORECAST and FORECAST.LINEAR?
FORECAST.LINEAR is the updated version in Excel 2016+ with identical functionality; Microsoft recommends FORECAST.LINEAR for new workbooks, but both formulas work identically.
How many historical data points do I need?
At least 2 points are required, but 10+ points provide much more reliable predictions by better establishing the trend line and reducing outlier impact.
Can I forecast multiple values at once?
Yes, use the TREND function as an array formula to predict multiple y-values simultaneously, or copy FORECAST across cells with different x inputs.

This was one task. ElyxAI handles hundreds.

Sign up