Standard Edition

Regression analysis with full diagnostics Simple, polynomial, multiple, and ANCOVA regression — with leverage plots, residual diagnostics, Cook’s D influence analysis, VIF, standardized betas, and prediction from the fitted model.

Regression as a process, not a single calculation

A regression result is only as good as the assumptions behind it. Influential observations can pull the line toward themselves. Correlated predictors inflate the standard errors until nothing looks significant. Non-constant variance makes the confidence intervals meaningless at one end of the range. The coefficients might look fine while the conclusions are wrong — and without diagnostics, you won’t know.

That’s why regression needs to be iterative: fit a model, examine whether the assumptions hold, identify what’s distorting the result, adjust, and re-fit. Not a single pass from data to p-value, but a cycle of building and checking until the model earns your confidence. Simple through multiple regression, ANOVA, ANCOVA, and logistic regression all work this way — the same diagnostic tools apply regardless of the model structure.

It is powerful enough so I can “play with the data”, which helps me to truly understand the results. Over the years I have tried several statistics programs, some Excel based, some not. But I always came back to Analyse-It, which is powerful enough for my purposes but the easiest to use.

Klaus T.
Chief R&D Officer
Pharmaceuticals

What’s included

Choose the model that fits the relationship

Linear, polynomial up to 6th order, logarithmic, exponential, power, and probit for simple regression. Multiple regression with any combination of continuous and categorical predictors, crossed terms, and interactions. Transform X or Y directly when a raw fit isn’t adequate. The scatter plot with fit line and confidence bands shows whether the model captures the pattern or forces a shape that isn’t there.

Spot the observations that are driving the result

The Outlier and Influence plot shows Cook’s D against Studentised Residuals for every observation — so you can see immediately whether a handful of unusual cases are distorting the model. Leverage plots isolate each predictor’s contribution after accounting for all others. Residual plots, sequence and lag plots reveal non-normality, heteroscedasticity, and autocorrelation. Save residuals, leverage, and Cook’s D back to the dataset for further investigation.

Everything you need to report and defend the model

Parameter estimates with confidence intervals. VIF to flag multicollinearity. Standardized betas to compare predictor importance on a common scale. R², adjusted R², AIC, and BIC for model comparison. Type I and Type III tests for each term. The model equation for documentation. Predicted against actual plot to see how well the model reproduces the data.

Predict from the fitted model and save the results

Predict Y for new X values directly from the model — no manual calculation from coefficients. Save fitted values, residuals, standardised and studentised residuals, leverage, and Cook’s D back to the dataset as new columns for further analysis or reporting.

Include categorical predictors with effect means and comparisons

Add categorical factors to the model and get adjusted effect means with confidence intervals for each level. Main effect and interaction plots show how the factors combine. Five multiple comparison procedures — Tukey-Kramer, Dunnett, Hsu, Scheffé, Student’s t — test specific differences between levels, just as in ANOVA.

Check whether the model form is adequate

The F-test for lack of fit tests whether a simple regression model captures the relationship or whether the data calls for a more complex form. If a linear fit shows significant lack of fit, try a polynomial or transform before reporting a result that doesn’t match the data.

Example analyses

See regression results in detail — parameter estimates, leverage plots, residual diagnostics, and influence analysis — using real datasets you can download and follow along with.

Simple regression — power fit
TV advertising vs retained impressions. 21 observations. Power regression with 95% confidence band, R², parameter estimates, residual diagnostics, outlier/leverage/influence plot with Cook’s D.

Multiple regression
Pulse rates, 8 predictors. 109 observations. Parameter estimates with VIF, Type III F-tests, leverage plots for all 8 predictors, residual diagnostics, outlier/influence plot.

Binary logistic regression
ICU patient survival, 17 predictors. 200 observations. Odds ratios with Wald 95% CIs, G² likelihood ratio test for model and each term.

Part of the Standard edition

Regression is one part of a complete statistical analysis toolkit. The Standard edition also includes ANOVA and ANCOVA, logistic regression, PCA and factor analysis, descriptive statistics, hypothesis testing, correlation, and categorical data analysis. See everything in the Standard edition →

Validated, reliable, trusted for over 30 years

Validated calculations Every statistic tested against NIST Standard Reference Datasets, published datasets and thousands of internal test-cases. No reliance on Excel'sbuilt-in functions. See how we develop and validate Analyse-it →

Data stays on your PC No cloud processing, no uploads, no third-party access. Your data never leaves your computer — essential when working with sensitive, confidential, or patient-identifiable data.

Standard Excel workbooks Analyses are ordinary Excel workbooks that you can share with colleagues, archive for audit, and open on any machine with Excel — no Analyse-it licence required.

No formulas to break Results contain no formulas, so they can't be accidentally edited or corrupted. The results you reported will be exactly what you find when you reopen the workbook.

Technical details

Linear fits

Simple linear regression
Polynomial regression (2nd to 6th order)
Logarithmic regression
Exponential regression
Power regression
Probit regression new in v5.50
Multiple linear regression
ANOVA new in v4.80
ANCOVA new in v4.80
Advanced models with simple, crossed, polynomial, and factorial terms
Transform X and Y variable new in v5.50

Other fits

Binary logistic regression

Model statistics

Model equation
R², adjusted R², AIC, BIC
Parameter estimates with CIs, VIF, standardized betas
F-test effect of model and each term
Type I (sequential) and Type III (partial) tests
Odds ratios with CIs (logistic)
Effect means for categorical variables new in v4.80
Main effect and interaction plots new in v4.80
Multiple comparisons of effect means new in v4.80

Diagnostics & plots

Scatter plot with fit line and confidence bands
Predicted against actual plot
Leverage plot for each term
Residual plots (raw, standardised)
Sequence and Lag-1 plots
Outlier and Influence plot (Cook’s D, Studentised Residuals)
F-test for lack of fit (simple models)

Prediction & save

Predict Y for X
Save fitted values, residuals, standardised residuals, studentised residuals, leverage, Cook’s D

Standard Edition Regression analysis with full diagnostics Simple, polynomial, multiple, and ANCOVA regression — with leverage plots, residual diagnostics, Cook’s D influence analysis, VIF, standardized betas, and prediction from the fitted model.