Point estimate calculator
For the sample 12, 15, 18, 21 and 24, the point estimate of the population mean is 18. Paste your sample and choose the mean, a proportion, the variance or the standard deviation to get one best guess and the formula behind it. Free in your browser.
- Free
- Runs in your browser
- No sign-up
- Your data stays on your device
Run — free
Worked example
- Values
- 12, 15, 18, 21, 24
- Parameter
- Mean
- N
- 5
- Point estimate
- 18
- Estimator
- sample_mean
A point estimate is a single number that stands in for an unknown population parameter. Introductory statistics courses treat the sample mean as the usual point estimate of the population mean, the sample proportion as the point estimate of a population proportion, and the unbiased sample variance or standard deviation as the point estimate of population spread. This calculator accepts a non-empty sample of finite numbers and an optional parameter choice, then returns that one best estimate together with the estimator name and the formula that produced it.
What a point estimate is and how each estimator is computed
A point estimate is a single number chosen as the best available stand-in for an unknown population parameter. In a first course in inferential statistics that parameter is usually the population mean mu, the population proportion p, the population variance sigma squared, or the population standard deviation sigma.
The corresponding sample statistics are the classical estimators: the sample mean x-bar equals the sum of the observations divided by n; the sample proportion p-hat equals the number of successes divided by n when the sample is coded as zeros and ones; the unbiased sample variance s squared equals the sum of squared deviations from the sample mean divided by n minus one; and the sample standard deviation s is the positive square root of that variance. Each of those formulas collapses a whole sample into one number.
When you leave the parameter field at its default, it reports the sample mean, which is the estimator students meet first and the one most homework problems mean when they ask for the point estimate of the population mean. When you switch the parameter, the same sample is reused and only the estimator changes. The implementation is exact arithmetic on ordinary floating-point numbers, with no sampling of its own and no random seed, so the same list always yields the same point estimate.
Difference from a descriptive mean-median-mode page matters for search: here the job is inference, one parameter, one number.
What you can do with it
Homework point-estimate checks
Students paste a sample and confirm the point estimate of μ (or p, σ², σ) against the estimator they used on the worksheet.
LMS and exam auto-grading
Mark point-estimate items with a deterministic oracle that rejects an empty sample and never invents a zero from no data.
Notebooks and process monitoring
Collapse a numeric sample to one best guess of a process mean or defect rate before you decide whether an interval is needed.
FAQ
What is a point estimate?
It is a single number used as the best guess of an unknown population parameter, such as the sample mean for μ or the sample proportion for p. This tool returns that one number, not an interval.
Which parameters can I estimate?
mean (default sample mean x̄ = Σxᵢ/n), proportion (p̂ from 0/1 trials), variance (unbiased s² with n−1), and stdev (s = √s²).
What happens if the sample values array is empty?
A point estimate requires at least one finite observation (two for variance and stdev). Empty input is not billed.
How is the sample variance computed?
As the unbiased estimator s² = Σ(xᵢ − x̄)² / (n − 1). That is the usual point estimate of the population variance; n must be at least two.
What does the point estimate calculator cost?
The browser tool is free and runs locally.
Related tools
For developers — API access
Everything on this page is available programmatically. This section is for teams who want to wire it into their own systems; everyone else can just use the tool above.
API endpoint
Prefer to automate it? One authenticated POST creates the task; the result comes back by webhook or a signed link. The same capability also runs here on the web, by email and from Telegram — and soon from our app too.
Call it from your stack
curl -X POST https://api.kit.forhosting.com/stat/point-estimate \
-H "Authorization: Bearer $KIT_KEY" \
-H "Content-Type: application/json" \
-d '{"values":[12,15,18,21,24]}'const res = await fetch("https://api.kit.forhosting.com/stat/point-estimate", {
method: "POST",
headers: {
"Authorization": `Bearer ${process.env.KIT_KEY}`,
"Content-Type": "application/json"
},
body: JSON.stringify({
"values": [
12,
15,
18,
21,
24
]
})
});
const { task_id } = await res.json();import os, requests
res = requests.post(
"https://api.kit.forhosting.com/stat/point-estimate",
headers={"Authorization": f"Bearer {os.environ['KIT_KEY']}"},
json={
"values": [
12,
15,
18,
21,
24
]
},
)
task_id = res.json()["task_id"]<?php
$res = file_get_contents("https://api.kit.forhosting.com/stat/point-estimate", false, stream_context_create([
"http" => [
"method" => "POST",
"header" => "Authorization: Bearer " . getenv("KIT_KEY") . "\r\nContent-Type: application/json",
"content" => '{"values":[12,15,18,21,24]}',
],
]));
$task = json_decode($res, true);body := bytes.NewBufferString(`{"values":[12,15,18,21,24]}`)
req, _ := http.NewRequest("POST", "https://api.kit.forhosting.com/stat/point-estimate", body)
req.Header.Set("Authorization", "Bearer "+os.Getenv("KIT_KEY"))
req.Header.Set("Content-Type", "application/json")
res, _ := http.DefaultClient.Do(req)Example request
{
"values": [
12,
15,
18,
21,
24
]
}Example response
{
"task_id": "tsk_a1b2c3d4e5f6a1b2c3d4e5f6",
"type": "stat.point_estimate",
"status": "queued",
"_links": {
"result": "/tasks/tsk_…/result"
}
}The API is asynchronous: the call returns a task_id immediately and the result arrives by webhook. Polling is capped at 1 req/s per task.
Pricing
Published price — no tokens, no invented credits. A failed task is never charged.
Limits
max_items | 100000 |
Errors
| HTTP | Code | Meaning |
|---|---|---|
401 | unauthorized | Missing or invalid API key. |
402 | insufficient_balance | Your balance doesn't cover the task price. |
404 | unknown_type | That task type doesn't exist. |
429 | rate_limited | Too many requests. Use the webhook instead of polling. |
Technical notes
An empty sample is rejected with invalid_input because there is no honest estimate from no observations.
What a point estimate is and how each estimator is computed
That is the point of a point estimate: you do not return a range or a p-value here, you return the single best guess the sample supports under the method-of-moments story taught in introductory texts. This calculator implements those four estimators on a finite numeric sample.
Input contract, response shape, and error behaviour
The parameter field is a closed list: mean, proportion, variance, or stdev, and it defaults to mean when omitted so a three-line homework check does not need extra ceremony. If you ask for a proportion, every entry must be exactly zero or one, because p-hat is defined on Bernoulli-coded trials; any other code is rejected instead of being silently recoded as a success. Variance and standard deviation need at least two observations so that the Bessel denominator n minus one is defined; a singleton sample is rejected for those two parameters. Send a JSON body with values, a required non-empty array of finite numbers that is the sample, and an optional parameter that selects which population quantity to estimate. Each sample entry may arrive as a JSON number or as a numeric string; missing entries, empty strings, NaN, and Infinity fail with invalid_input and name the offending index so a form can highlight it. On success the body always carries the same five keys: parameter, n, the numeric point_estimate, the estimator name, and a short formula string you can print next to the answer. Example: values twelve, fifteen, eighteen, twenty-one, twenty-four with parameter mean yields n equals five and point_estimate equals eighteen, because the sum is ninety and ninety divided by five is eighteen. Pair this endpoint with a confidence-interval tool if you later need a range; this capability returns only the point.
Where a point estimate calculator earns its keep
Classrooms and tutoring platforms use a point estimate calculator so learners can check the sample mean, the sample proportion, or the unbiased sample variance against the method they just wrote on paper, without mixing those answers into a descriptive dump of median, mode, and range. Failed validation does not consume balance. Exam-prep generators need a deterministic oracle that marks an empty sample as invalid_input instead of silently returning zero, because a grader that treats an empty success payload as a correct zero will pass students who submitted nothing. Lab notebooks and quality scripts often need one number — the current best guess of a process mean or a defect rate — before they decide whether to compute an interval or stop a line; calling a tiny parse endpoint avoids pulling a full statistics library for a one-scan average.