Linear Regression Calculator

Calculate linear regression equations, correlation coefficients, and scatter plots online, with support for manual input, CSV import, and batch data analysis.

Bug report

Browser execution mode: Your data is processed in your browser and is not uploaded to the server.

Speed and Stability: Processing speed depends on your device and browser. For large batch work, the desktop version may be more stable.

Loading tool, please wait...

If the online tool fails to load or run, try the desktop tool.https://tools.yikeaigc.com/

Tool Usage

Return to old version

Prepare X/Y data, set model parameters, then calculate and export results.

Supports manual entry, pasted tables, CSV/TSV/TXT files, and sample data. Batch upload processes all files; only the first 20 are shown for review.

OLS fit R² / correlation Residuals & outliers Report / CSV / charts Save settings

Batch files No file selected

After uploading multiple CSV, TSV, or TXT files, the first 20 file names will appear here. Batch calculation generates all reports and downloads them as a ZIP.

Input data

Choose a data source. Each row needs one X and Y value. Headers are allowed. Commas, tabs, spaces, and semicolons are auto-detected.

Data table Linear regression needs at least 2 points; polynomial regression needs more points than the degree

No.	X value	Y value	Action

Paste X/Y data Supports comma, space, Tab, semicolon delimiters

Bulk data

Upload data file Select multiple files; first 20 shown

Click to choose files, or drag them here Supports .csv, .tsv, .txt. First column is X, second is Y. Headers allowed.

Sample data Load and calculate directly

Set model

Try linear regression first, then use the scatter plot, R², and residuals to decide if a polynomial fit is needed.

Regression modelHigher polynomial degrees increase overfitting risk

Model type

Custom degree

Decimal places

Confidence level

Outlier threshold

Predict X Values

Display & DiagnosticsSelect items to include in results and reports

Scatter Plot & Fit Line Confidence Band Residual Analysis Outlier Detection Descriptive Statistics Compare 1st-3rd Order Models

Calculate & export

After calculation, view the equation, metrics, charts, residuals, and predicted values. If multiple files are uploaded, batch results will also be generated.

Export ResultsComplete the calculation before downloading reports or charts

Set data and model first, then click “Start Regression Analysis”.

Scatter Plot & Fit Line

Waiting for ResultsSet data and model first to generate a regression chart.

Regression Equation-

Correlation Coefficient r-

Coefficient of Determination R²-

Standard Error SE-

Key Statistics

Coefficients & Intervals

Prediction Results

Descriptive Statistics

Residual Analysis

Model Comparison

Batch Results Preview

Instructions

Software Usage Instructions

Input Data: You can choose to enter data manually, paste data, or upload a CSV, TSV, or TXT file. Enter one set of X and Y values per line. The file may include a header row.
Select Model: Supports linear regression, quadratic polynomial, cubic polynomial, and custom degree. It is recommended to start with linear regression, then use the scatter plot and residuals to determine whether curve fitting is needed.
Set Parameters: You can set decimal places, confidence level, and outlier threshold, and choose whether to display the scatter plot, confidence band, residual analysis, outlier detection, descriptive statistics, and model comparison.
Enter Prediction Values: If you need to predict Y values, enter one or more X values in “Prediction X Values,” separated by commas.
Start Calculation: Click “Start Regression Analysis,” and the system will generate the regression equation, correlation coefficient, R², adjusted R², standard error, F statistic, p-value, residuals, and charts.
Export Results: After the calculation is complete, you can copy the equation, download a TXT report or CSV data, and export PNG or SVG charts; batch files will generate a ZIP results package.

FAQ

A: Linear regression requires at least 2 valid data points, and polynomial regression requires the number of data points to be greater than the model degree. The more sufficient the data, the more stable the results usually are.

A: The correlation coefficient r describes the direction and strength of the linear relationship between X and Y; R² represents the proportion of variation in Y explained by the model. In linear regression, R² is usually related to the square of r.

A: When the scatter plot shows a clear curved trend, or the residual plot shows a regular curvature pattern, you can try quadratic or cubic polynomial regression. The degree should not be too high to avoid overfitting the existing data.

A: A residual is the difference between the actual Y value and the predicted Y value. Observing residuals can help determine whether the model is appropriate and can also identify data points with large deviations.

A: CSV, TSV, or TXT files should contain two columns: the first column for X values and the second column for Y values. Headers are supported, as are comma, Tab, space, and semicolon delimiters.

A: The mean confidence interval is used to estimate the range of the model’s average response, while the single-point prediction interval is used to estimate the range of a single new observation. The latter is usually wider.