Statistics you can defend in front of a reviewer.
Upload a dataset from any field. Celeus suggests the right statistical method and tells you why, runs it deterministically in R, and interprets the result - so the number in your paper is the number your data actually supports.
Paired t-test
Data: Student (1908), Biometrika 6(1):1-25
| Celeus | p = 0.00283289 | |
|---|---|---|
| Independent R computation | p = 0.00283289 | matches Celeus |
Same number, computed two independent ways. Every run ships its exact R script, seed and package versions. See all checks
See a run from upload to package
Transcript
Celeus: from a dataset to a reproducible result.
We drop in a synthetic dataset: forty rows, five columns.
Celeus scans every column first. Two of them, patient name and MRN, look like direct identifiers.
The review recommends replacing both columns with coded stand-ins. We apply the recommendation as proposed.
That creates a new, de-identified version. The original upload is kept, and a fresh scan finds no identifiers.
Next, Celeus ranks suitable analyses from the de-identified profile, without calling an AI model. To compare the outcome between the two arms, it recommends Welch's t-test.
The test, outcome and group come pre-filled. We add a research question, which is saved with the results.
The analysis runs in R, the statistical programming language, with a fixed seed. The wait is shortened in this video.
The result: the treatment group's mean is 6.7 points higher, with a 95 percent confidence interval of 4.25 to 9.15, and p below 0.001.
Below it is a plain-language summary written without AI, and a warning: with forty rows, only large effects can be detected reliably.
An AI interpretation is added on top, clearly labelled. Only de-identified metadata reaches the model, and the computed numbers remain the ground truth.
Here is the exact R code that reproduces this result, with its seed and the engine version.
We check the results and confirm them right here, next to the download. The confirmation is saved with the run, and the reproducible package unlocks: the data, the script and the pinned package versions, together.
How it works
1
Upload
CSV, TSV or Excel on every plan; SPSS, Stata and SAS on paid plans. Or add one of the public sample datasets to try it first.
2
Method chosen and explained
Celeus suggests a method for your question and your data, grades the alternatives it considered, and checks the assumptions. If you pick something else, that choice is recorded.
3
Deterministic run
Seeded, in R, on pinned package versions. Run it again on the same engine and you get the same numbers.
4
Interpretation and flags
A plain-language reading of the result, with failed assumptions and other problems flagged where you will see them.
5
Reproducible package
The plan, the exact R script, the seed, session info, the results and the audit log, in one download.
Works with any research data
The subject of your data does not matter. What matters is its shape and your question.
| Question | A method that fits |
|---|---|
| Clinical researchDo patients on the new treatment stay event-free longer?Kaplan-Meier and Cox regression | Kaplan-Meier and Cox regression |
| PsychologyDid scores change after the intervention?Paired t-test or Wilcoxon signed-rank | Paired t-test or Wilcoxon signed-rank |
| EcologyDo species counts differ between sites?Poisson or negative-binomial regression | Poisson or negative-binomial regression |
| EducationDo outcomes differ across three schools?One-way ANOVA or Kruskal-Wallis | One-way ANOVA or Kruskal-Wallis |
| Public healthIs an exposure associated with the outcome?Chi-squared test or logistic regression | Chi-squared test or logistic regression |
| Lab scienceDo two assays agree?Bland-Altman agreement | Bland-Altman agreement |
What reaches the AI, and what never does
Every path from your data to the AI model passes a guard that refuses data tables. Every upload is scanned for identifiers, and columns flagged as identifiers, free text or dates are excluded from the AI by default. You decide what happens to each flagged column.
Reaches the AI
- Column names, types and variable labels stored in your file
- Summary statistics, including each numeric column's minimum and maximum
- Category names, with counts of 11 or fewer not listed
- Aggregate results of your analyses
- Text you write: chat messages, research questions, notes and codebook descriptions
Never reaches the AI
- Your uploaded file
- Data rows or data tables
Text you write is sent as you wrote it, so keep identifiers out of it. That matters most for patient records, and just as much for survey responses, student records or anything else collected under a confidentiality promise.How Celeus handles your data
More than a test runner
Chat with your data
Ask about your dataset and results in plain language. The assistant works from the de-identified profile and aggregate results, never from data rows.
Draft methods and results
Turn a run into Methods, Results and a structured abstract. Every number comes from the run itself, not from the model. Export to Word, LaTeX or Markdown.
Share with collaborators
Send a time-limited, read-only link to a report snapshot, or give a named researcher in another workspace view or run access to a project.
Sensitivity analysis
Compare the complete-case result with multiple imputation for the same analysis, and see whether the two agree.
Censored and below-detection data
Values below a detection limit are modelled as censored (Peto-Peto test, Tobit-style regression), never replaced with a guessed value.
REDCap data dictionaries
Import a REDCap data dictionary to label your variables and categories. You review each change before it is applied.
36statistical methods, checked in the open
Every plan gets every method. On well-known public datasets, each result is checked against an independent computation, and a subset against values published in the literature. The report is generated, not written by hand.
Read the validation reportComparing groups
- Welch's t-test
- Student's t-test
- Paired t-test
- Mann-Whitney U test
- Wilcoxon signed-rank test
- One-way ANOVA
- Welch's ANOVA
- Kruskal-Wallis test
- Repeated-measures ANOVA (Greenhouse-Geisser)
- ANCOVA (analysis of covariance)
- Two-way ANOVA (factorial)
Association and correlation
- Chi-squared test of association
- Fisher's exact test
- Pearson correlation
- Spearman correlation
- McNemar's test
Regression
- Linear regression
- Logistic regression
- Firth penalised logistic regression
- Poisson regression
- Negative-binomial regression
- Ordinal logistic regression
- Multinomial logistic regression
Mixed-effects models
- Linear mixed-effects model (random intercept)
- Mixed-effects logistic regression (random intercept)
- Mixed-effects Poisson regression (random intercept)
Survival and censored data
- Peto-Peto generalized Wilcoxon test
- Censored regression (Tobit-style)
- Kaplan-Meier (log-rank test)
- Cox proportional-hazards regression
- Weibull accelerated failure time regression
Diagnostic accuracy and agreement
- ROC curve / AUC
- Diagnostic accuracy (2x2)
- Bland-Altman agreement
- Compare two ROC curves (DeLong)
Descriptive
- Descriptive table (Table 1)
Pricing
One researcher or a whole lab - the per-person price stays the same.
Free
See it work on your own data.
$0/mo
- 10 analysis runs a month
- ~15 AI actions a month
- 100 MB storage, 1 person
- All 36 statistical methods
- CSV, TSV and Excel import
- Identifier scan and de-identification
- Reproducible package on every run
Pro
One researcher's full toolkit.
$19/mo
- 500 analysis runs a month
- ~300 AI actions a month
- 5 GB storage
- Shareable report links
- Multiple reports per project
- SPSS, Stata and SAS import
- Advanced data preparation
- Assistant without a dataset
- Share a project with a collaborator
- Everything in Free
Built for labs
Team
A lab. Same price per person, plus the lab layer.
$19/mo per seat
- Per seat: 500 runs and ~150 AI actions a month, 2.5 GB storage, pooled across the lab
- Shared workspace - the lab owns the data
- Admin-managed membership and billing
- Everything in Pro
Enterprise
The org wrapper around Team.
Contact us
- Everything in Team
- Need more than 25 seats or custom terms? Contact us.
Need more tokens or runs without changing plans? Top-up packs: AI pack $15 for 1M tokens, Analysis pack $12 for 200 runs.
Out of storage? A storage top-up adds 25 GB added permanently, on any plan - no plan change and no seat change needed. Buy it from Settings > Billing; the price is shown before you pay.
Every plan above starts by signing in. If that does not let you through, contact support and we will follow up by email.
Celeus is for research questions on research data. It does not make diagnosis or treatment decisions about individual patients.