{ "openapi": "3.0.1", "info": { "title": "Statistics Calculator: A/B Test, t-Test, ANOVA, Regression", "description": "A/B test significance, t-tests, ANOVA, chi-square, correlation, linear and logistic regression, sample size and more, with plain-language results, effect sizes, assumption checks and charts. From numbers, CSV, Excel, Google Sheets or an Apify dataset. Checked against R. For AI agents.", "version": "0.1", "x-build-id": "MAdGYtHrcW3EqJqGf" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/jumblemog~statistics-calculator/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-jumblemog-statistics-calculator", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for its completion, and returns Actor's dataset items in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/jumblemog~statistics-calculator/runs": { "post": { "operationId": "runs-sync-jumblemog-statistics-calculator", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/jumblemog~statistics-calculator/run-sync": { "post": { "operationId": "run-sync-jumblemog-statistics-calculator", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } } }, "components": { "schemas": { "inputSchema": { "type": "object", "properties": { "analysis": { "title": "Analysis", "enum": [ "auto", "abTest", "compareGroups", "paired", "oneSample", "correlation", "regression", "crosstab", "describe", "normality", "outliers", "sampleSize" ], "type": "string", "description": "What to run. Auto picks from the data: variants give an A/B test, groups a group comparison, a table of counts a chi-square test, values descriptive statistics (a one-sample test with Test value), and baselineRate alone a sample size. For rows, the columns named decide: predictors give a regression, Group column a group comparison (an A/B test for a yes/no outcome), Row and Column fields a crosstab, two or more columns a correlation.", "default": "auto" }, "analyses": { "title": "Several analyses on the same data", "type": "array", "description": "Runs each of these on the same data, such as describe, normality and correlation. Each analysis is one charge.", "items": { "type": "string", "enum": [ "abTest", "compareGroups", "paired", "oneSample", "correlation", "regression", "crosstab", "describe", "normality", "outliers", "sampleSize" ], "enumTitles": [ "A/B test (conversion rates, z-test or Fisher, Bayesian)", "Compare groups (t-test, ANOVA, Mann-Whitney, Kruskal-Wallis)", "Paired or before/after (paired t-test, Wilcoxon, repeated measures)", "One sample vs a target (one-sample t-test, binomial test)", "Correlation (Pearson, Spearman, Kendall, matrix)", "Regression (linear or logistic)", "Crosstab (chi-square, Fisher's exact, goodness of fit)", "Descriptive statistics", "Normality test (Shapiro-Wilk, D'Agostino)", "Outliers (IQR, z-score, modified z, Grubbs)", "Sample size and power (A/B test planning)" ] } }, "tasks": { "title": "Tasks (JSON)", "type": "array", "description": "Many analyses in one run, each an object with its analysis, data and settings: [{\"analysis\": \"abTest\", \"variants\": [...]}, {\"analysis\": \"t-test\", \"groups\": {...}}]. Settings outside tasks apply to every task, and a task with its own data doesn't use the shared data. Common names such as t-test, anova, mann-whitney, chi-square, pearson, logistic or power work too. Up to 500 tasks, one charge each." }, "variants": { "title": "A/B test counts", "type": "array", "description": "Visitors and conversions per variant, control first: [{\"name\": \"A\", \"visitors\": 4800, \"conversions\": 192}, ...]. Users, sessions, trials or n work for visitors; successes, orders, clicks or signups for conversions; a rate works too. One entry with Test value runs a one-sample test of the rate." }, "groups": { "title": "Groups of numbers", "type": "object", "description": "Each group's values: {\"Control\": [12.1, 9.8, 11.4, ...], \"Treatment\": [14.2, 13.1, ...]}. Two groups give a t-test (or Mann-Whitney U), three or more an ANOVA (or Kruskal-Wallis) with pairwise comparisons. For a paired analysis, equal-length lists are matched by position." }, "values": { "title": "Values (one list)", "type": "array", "description": "One list of numbers, such as [31, 28, 35, 30, 41]: descriptive statistics, a one-sample test against Test value, a normality test or outliers." }, "table": { "title": "Table of counts", "type": "array", "description": "Counts for a chi-square or Fisher's exact test, rows with a header: [[\"\", \"Bought\", \"Didn't buy\"], [\"Women\", 30, 10], [\"Men\", 20, 25]], or plain [[30, 10], [20, 25]]. One list of counts such as [18, 22, 30, 30] tests goodness of fit against an even split (or Expected split)." }, "data": { "title": "Data rows (JSON)", "type": "array", "description": "Rows as objects, such as [{\"variant\": \"A\", \"revenue\": 12.5}, ...], or the items of any API. Name the columns below, or leave them empty to have them found." }, "csv": { "title": "CSV data", "type": "string", "description": "CSV or TSV text with a header row, such as group,score on the first line and A,12.5 on the next. Comma, semicolon, tab and pipe separators work." }, "dataUrl": { "title": "Data file link", "type": "string", "description": "A link to a CSV, TSV, JSON or Excel (.xlsx) file. Google Sheets share links (shared as \"Anyone with the link\"), GitHub file links and JSON API answers work. The site's robots.txt is respected." }, "datasetId": { "title": "Apify dataset", "type": "string", "description": "Analyze the items of a dataset in your account, such as a scraper's results (prices, ratings, review counts). Up to 500,000 items; the tool gets read access to this dataset only." }, "outcome": { "title": "Outcome column", "type": "string", "description": "The measured or predicted column, such as revenue, score or converted. With Group column, a yes/no outcome gives an A/B test and numbers a group comparison; with predictors, a regression." }, "groupField": { "title": "Group column", "type": "string", "description": "The column that splits the rows into groups, such as variant, group, treatment or country." }, "predictors": { "title": "Predictor columns", "type": "array", "description": "Columns that predict the outcome in a regression. Number columns are used as they are; text columns become categories compared with their most common level (or Reference level).", "items": { "type": "string" } }, "columns": { "title": "Columns to analyze", "type": "array", "description": "Number columns: two for a correlation (three or more for a correlation matrix), any for descriptive statistics, normality or outliers, or one column per group (wide data) for a group comparison or a paired test. Empty: all number columns.", "items": { "type": "string" } }, "subjectField": { "title": "Subject column (paired data)", "type": "string", "description": "For paired data in long form: the column that identifies each subject (person, store, page) measured once in each condition of Group column." }, "rowField": { "title": "Crosstab row column", "type": "string", "description": "The first category column of a crosstab, such as gender." }, "columnField": { "title": "Crosstab column column", "type": "string", "description": "The second category column of a crosstab, such as purchased." }, "countField": { "title": "Count column", "type": "string", "description": "For rows that are already counted (one row per combination): the column with the counts." }, "categoricalColumns": { "title": "Number columns to treat as categories", "type": "array", "description": "Regression predictors that hold numbers but are categories, such as a cylinder count, a year or a store number.", "items": { "type": "string" } }, "labelField": { "title": "Label column", "type": "string", "description": "A column that names each row (product, name, ID), shown with each outlier." }, "confidenceLevel": { "title": "Confidence level (%)", "minimum": 50, "maximum": 99.99, "type": "number", "description": "The level of confidence intervals and the significance threshold: 95 means a p-value below 0.05 is significant.", "default": 95 }, "alpha": { "title": "Significance level (alpha)", "minimum": 0.0001, "maximum": 0.5, "type": "number", "description": "Instead of a confidence level: the p-value threshold, such as 0.05 or 0.01." }, "alternative": { "title": "Hypothesis", "enum": [ "two-sided", "greater", "less" ], "type": "string", "description": "Two-sided looks for a difference either way. Greater and less test one direction: the variant, the second group or condition, or the data above (or below) the control, the first group or the test value; for correlations, a positive (or negative) correlation.", "default": "two-sided" }, "method": { "title": "Test family", "enum": [ "auto", "parametric", "nonparametric" ], "type": "string", "description": "Auto uses t-tests and ANOVA unless a group with fewer than 30 values fails a normality test, then the rank-based test. Both results are always reported.", "default": "auto" }, "equalVariances": { "title": "Assume equal variances", "type": "boolean", "description": "Off (recommended): Welch's t-test and Welch's ANOVA, which don't assume equal spreads. On: Student's t-test and the classic ANOVA with Tukey's HSD.", "default": false }, "correction": { "title": "Multiple-comparison correction", "enum": [ "holm", "bonferroni", "fdr_bh", "none" ], "type": "string", "description": "Adjusts p-values when several variants, group pairs or correlations are tested at once.", "default": "holm" }, "control": { "title": "Control group", "type": "string", "description": "The name of the control or baseline variant or group. Empty: one named control, baseline, original, A..., or else the first." }, "testValue": { "title": "Test value", "type": "number", "description": "For a one-sample test: the target or benchmark the data are tested against, such as 30 (minutes) or, for a rate, 5 or 0.05 (5%)." }, "correlationMethod": { "title": "Correlation method", "enum": [ "pearson", "spearman", "kendall" ], "type": "string", "description": "For correlations. All three are reported for a pair of columns.", "default": "pearson" }, "regressionType": { "title": "Regression type", "enum": [ "auto", "linear", "logistic" ], "type": "string", "description": "Linear regression for a number outcome, logistic regression for a yes/no outcome.", "default": "auto" }, "robustErrors": { "title": "Robust standard errors", "enum": [ "auto", "always", "never" ], "type": "string", "description": "Heteroscedasticity-robust (HC3) standard errors for linear regression. Auto switches them on when the Breusch-Pagan test finds the residuals' spread changing.", "default": "auto" }, "reference": { "title": "Reference level", "type": "string", "description": "For category predictors: the level the others are compared with. Empty: each column's most common level." }, "predict": { "title": "Predict for", "type": "array", "description": "New rows to predict with the fitted model, each with a value for every predictor: [{\"ad_spend\": 1200, \"region\": \"West\"}]. Linear models give the prediction with its confidence and prediction intervals; logistic models the probability with its interval." }, "baselineRate": { "title": "Baseline conversion rate (%)", "minimum": 0, "maximum": 100, "type": "number", "description": "The control's current conversion rate, such as 4 (or 0.04) for 4%. Values below 1 are read as a fraction, so write 0.5% as 0.005." }, "minimumDetectableEffect": { "title": "Minimum detectable effect", "minimum": 0, "type": "number", "description": "The smallest lift worth detecting: a relative lift in % (10 means 4% to 4.4%), or with Effect type absolute, percentage points (or units for averages). Empty with Visitors per variant: the lift those visitors can detect." }, "mdeType": { "title": "Effect type", "enum": [ "relative", "absolute" ], "type": "string", "description": "How Minimum detectable effect is meant.", "default": "relative" }, "power": { "title": "Power (%)", "minimum": 50, "maximum": 99.9, "type": "number", "description": "The chance of detecting the effect if it is real; 80 is usual, 90 for important tests.", "default": 80 }, "numberOfVariants": { "title": "Number of variants", "minimum": 2, "maximum": 20, "type": "integer", "description": "Variants in the test including the control; with more than two, each comparison is held to a Bonferroni-corrected significance level.", "default": 2 }, "dailyTraffic": { "title": "Visitors per day", "minimum": 0, "type": "number", "description": "Visitors entering the test per day, across all variants, to estimate how many days it takes." }, "sampleSizePerVariant": { "title": "Visitors per variant", "minimum": 2, "type": "integer", "description": "Instead of a minimum detectable effect: the visitors you have per variant, to find the smallest lift they can detect." }, "standardDeviation": { "title": "Standard deviation (averages)", "minimum": 0, "type": "number", "description": "For tests of averages (revenue per visitor, time on page): the metric's standard deviation, with Minimum detectable effect in the same units and Effect type absolute." }, "effectSize": { "title": "Effect size (Cohen's d)", "minimum": 0, "type": "number", "description": "For tests of averages, instead of a standard deviation: the standardized difference to detect (0.2 small, 0.5 medium, 0.8 large)." }, "expectedSplit": { "title": "Expected split", "type": "array", "description": "The planned traffic split for the sample ratio mismatch check, such as [50, 50] or [90, 10], in the order of the variants. For a goodness-of-fit test: the expected share of each category." }, "outlierMethod": { "title": "Outlier rule", "enum": [ "iqr", "zscore", "modifiedZ", "grubbs" ], "type": "string", "description": "How outliers are found.", "default": "iqr" }, "outlierThreshold": { "title": "Outlier threshold", "minimum": 0, "type": "number", "description": "The rule's cut-off: the IQR multiple (default 1.5; 3 for extreme outliers), or the z-score or modified z-score (defaults 3 and 3.5)." }, "chart": { "title": "Charts", "type": "boolean", "description": "Draws a chart for each analysis: rates with intervals, box plots, scatter plots with the fit, heatmaps, histograms, Q-Q plots or sample-size curves.", "default": true }, "chartFormats": { "title": "Chart formats", "type": "array", "description": "PNG for chats, emails, docs and slides; SVG and PDF stay sharp at any size; HTML shows values on hover.", "items": { "type": "string", "enum": [ "png", "svg", "pdf", "html" ], "enumTitles": [ "PNG", "SVG", "PDF", "Interactive HTML (hover for values)" ] }, "default": [ "png" ] }, "chartTheme": { "title": "Chart theme", "enum": [ "light", "dark" ], "type": "string", "description": "Light or dark background.", "default": "light" }, "title": { "title": "Title", "type": "string", "description": "A title for the result and its chart. Empty: one made from the analysis and the data." }, "valueLabel": { "title": "Value name", "type": "string", "description": "What the numbers measure, for sentences and axis titles, such as order value or minutes." } } }, "runsResponseSchema": { "type": "object", "properties": { "data": { "type": "object", "properties": { "id": { "type": "string" }, "actId": { "type": "string" }, "userId": { "type": "string" }, "startedAt": { "type": "string", "format": "date-time", "example": "2025-01-08T00:00:00.000Z" }, "finishedAt": { "type": "string", "format": "date-time", "example": "2025-01-08T00:00:00.000Z" }, "status": { "type": "string", "example": "READY" }, "meta": { "type": "object", "properties": { "origin": { "type": "string", "example": "API" }, "userAgent": { "type": "string" } } }, "stats": { "type": "object", "properties": { "inputBodyLen": { "type": "integer", "example": 2000 }, "rebootCount": { "type": "integer", "example": 0 }, "restartCount": { "type": "integer", "example": 0 }, "resurrectCount": { "type": "integer", "example": 0 }, "computeUnits": { "type": "integer", "example": 0 } } }, "options": { "type": "object", "properties": { "build": { "type": "string", "example": "latest" }, "timeoutSecs": { "type": "integer", "example": 300 }, "memoryMbytes": { "type": "integer", "example": 1024 }, "diskMbytes": { "type": "integer", "example": 2048 } } }, "buildId": { "type": "string" }, "defaultKeyValueStoreId": { "type": "string" }, "defaultDatasetId": { "type": "string" }, "defaultRequestQueueId": { "type": "string" }, "buildNumber": { "type": "string", "example": "1.0.0" }, "containerUrl": { "type": "string" }, "usage": { "type": "object", "properties": { "ACTOR_COMPUTE_UNITS": { "type": "integer", "example": 0 }, "DATASET_READS": { "type": "integer", "example": 0 }, "DATASET_WRITES": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_READS": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_WRITES": { "type": "integer", "example": 1 }, "KEY_VALUE_STORE_LISTS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_READS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_WRITES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_INTERNAL_GBYTES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_EXTERNAL_GBYTES": { "type": "integer", "example": 0 }, "PROXY_RESIDENTIAL_TRANSFER_GBYTES": { "type": "integer", "example": 0 }, "PROXY_SERPS": { "type": "integer", "example": 0 }, "PROXY_UNBLOCKER_UNITS": { "type": "integer", "example": 0 } } }, "usageTotalUsd": { "type": "number", "example": 0.00005 }, "usageUsd": { "type": "object", "properties": { "ACTOR_COMPUTE_UNITS": { "type": "integer", "example": 0 }, "DATASET_READS": { "type": "integer", "example": 0 }, "DATASET_WRITES": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_READS": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_WRITES": { "type": "number", "example": 0.00005 }, "KEY_VALUE_STORE_LISTS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_READS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_WRITES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_INTERNAL_GBYTES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_EXTERNAL_GBYTES": { "type": "integer", "example": 0 }, "PROXY_RESIDENTIAL_TRANSFER_GBYTES": { "type": "integer", "example": 0 }, "PROXY_SERPS": { "type": "integer", "example": 0 }, "PROXY_UNBLOCKER_UNITS": { "type": "integer", "example": 0 } } } } } } } } } }