Normal Distribution Calculator — PDF, CDF & Inverse CDF

Guide

About the Normal Distribution Calculator

Last updated: July 2026 · Reviewed by the NeftCal editorial team

This free normal distribution calculator — also a bell curve calculator, Gaussian distribution calculator, and cumulative distribution function calculator — computes the probability density (PDF), the cumulative probability (CDF), and the inverse CDF for any distribution defined by a mean μ and standard deviation σ. Enter a value to find how much of the bell curve lies below it, or enter a percentile to find the value that corresponds to it — useful for anything from test-score cutoffs to manufacturing tolerances to financial risk modeling.

What This Calculator Measures

The normal distribution is a symmetric, bell-shaped curve defined entirely by its mean (center) and standard deviation (spread). This calculator's Value → Probability tab measures the density at a point and the cumulative probability P(X ≤ x) — the share of the distribution lying at or below that value. Its Probability → Value tab does the reverse: given a target probability, it finds the value on the distribution's scale that corresponds to it.

Who Should Use This Calculator

This tool is for students and instructors working through probability density function calculator and cumulative distribution function calculator problems, data scientists and analysts modeling continuous variables, quality-control engineers setting tolerance limits, finance professionals estimating value-at-risk under a normal assumption, and researchers computing confidence intervals or critical values from a Gaussian distribution.

Why the Normal Distribution Matters

The normal distribution is the most widely used continuous probability distribution in statistics because, per the central limit theorem, the sum or average of many independent random effects tends toward a normal shape — regardless of the shape of the underlying individual effects. This makes it a natural default model for measurement error, natural variation in biological traits, and aggregated data of all kinds.

Real-World Applications

Standardized testing programs use the normal distribution to set percentile-based cutoffs and score scales. Manufacturers assume normally distributed measurement error when setting Six Sigma quality-control limits. Financial analysts use it (with caveats) to estimate value-at-risk and the probability of extreme portfolio moves. Biologists and doctors use it to model natural variation in height, blood pressure, and other continuous biological measurements. Researchers rely on the standard normal distribution to compute confidence intervals and critical values in hypothesis testing.

Tips for Accurate Results

Confirm your data is reasonably close to normally distributed before trusting the probability results — this calculator assumes true normality, and real data can be skewed.
Remember the CDF has no simple closed-form algebraic formula — this calculator (like all statistical software) uses a numerical polynomial approximation, accurate to many decimal places but not an exact fraction.
Use the empirical (68-95-99.7) rule as a quick sanity check on any Value → Probability result before trusting the exact figure.
In the Probability → Value tab, enter a probability strictly between 0 and 1 — exactly 0 or 1 corresponds to −∞ or +∞ and has no finite answer.

Formula

The Normal Distribution Formulas, Explained

How this normal distribution calculator computes density, cumulative probability, and the inverse CDF

Z-Score (standardization)

z = (x − μ) ÷ σ

Probability Density Function (PDF)

f(x) = [1 ÷ (σ√(2π))] · e^(−(x−μ)² ÷ (2σ²))

Cumulative Distribution Function (CDF)

P(X ≤ x) = Φ(z) — no elementary closed-form formula; computed via the Abramowitz & Stegun polynomial approximation to the standard normal CDF

Inverse CDF (value at a percentile)

x = μ + z·σ, where z = Φ⁻¹(p) is found via the Beasley-Springer-Moro algorithm

Where:
μ (mu) = the mean of the distribution.
σ (sigma) = the standard deviation of the distribution (must be greater than 0).
x = a value on the distribution's original scale.
z = the standardized z-score for x.
Φ (Phi) = the standard normal CDF; Φ⁻¹ is its inverse.
p = a probability/percentile strictly between 0 and 1, used in the inverse direction.

📈

PDF is a Density, Not a Probability

The PDF value at a point can exceed 1 for a small σ — only the area under the curve between two points is an actual probability.

∫

The CDF Has No Closed Form

P(X ≤ x) can only be written as an integral, or in terms of the error function (erf) — it's always computed numerically, never algebraically.

⚖️

Symmetry Around the Mean

A probability of exactly 0.5 in the inverse direction always returns x = μ, since the bell curve is perfectly symmetric.

⚙️ Why This Formula Works

The PDF's exponential term concentrates probability density near the mean and tapers it symmetrically in both directions, producing the familiar bell shape. Because no elementary function integrates the Gaussian exponential exactly, the CDF is instead approximated with a high-precision polynomial (Abramowitz & Stegun) that matches the true integral to within about 10⁻⁷ — accurate enough for essentially any practical statistical use, even though it is not an exact algebraic expression.

🎯 When to Use Each Mode

Value → Probability: you have a value and want its density, cumulative probability, or upper-tail probability
Probability → Value: you have a target percentile (e.g. the 90th) and want the corresponding value

📋 Assumptions

The standard deviation σ is greater than 0
The underlying data genuinely follows, or closely approximates, a normal distribution
The probability p entered in the inverse direction is strictly between 0 and 1

⚠️ Limitations of the Formula

The CDF and inverse CDF are numerical approximations, not exact closed-form values — though accurate to many decimal places
Undefined when σ = 0, or when p is exactly 0 or 1 in the inverse direction
Real-world data that isn't actually normally distributed (skewed, bounded, or multimodal) will not match these results

Walkthrough

Step-by-Step: How to Use the Normal Distribution Calculator

From picking a direction to interpreting the probability

Pick a mode

Choose "Value → Probability" to find the density and cumulative probability at a value, or "Probability → Value" to find the value at a given percentile.

Enter the mean and standard deviation

Enter μ (mean) and σ (standard deviation) describing your normal distribution — these define its center and spread.

Enter a value or a probability

For Value → Probability, enter x. For Probability → Value, enter a probability strictly between 0 and 1 (e.g. 0.9 for the 90th percentile).

Click "Calculate"

The calculator converts your input to a z-score internally, then evaluates the PDF formula and the standard normal CDF approximation (or its inverse).

Read the result

Value → Probability shows P(X ≤ x), P(X > x), the density at x, and the z-score. Probability → Value shows the corresponding value and its z-score.

Verify the answer

Check that P(X ≤ x) + P(X > x) = 1, or confirm that a probability of exactly 0.5 in Probability → Value returns the mean.

Example

Worked Example

An IQ-style distribution with mean 100 and standard deviation 15

Scenario

A distribution has mean μ = 100 and standard deviation σ = 15. What proportion of the distribution falls at or below x = 115? And what value corresponds to the 90th percentile?

Mean (μ)100

Std Dev (σ)15

Value (x)115

Step 1 — Convert x to a z-score: z = (115 − 100) ÷ 15 = 1.0.

Step 2 — Evaluate the CDF: Φ(1.0) ≈ 0.8413, so P(X ≤ 115) ≈ 84.13%.

Step 3 — Find the upper tail: P(X > 115) = 1 − 0.8413 = 15.87%.

Step 4 — Evaluate the PDF at x = 115: f(115) = [1 ÷ (15√(2π))] · e^(−1²/2) ≈ 0.016131 — a density, not a probability.

Step 5 — Reverse direction, find the 90th percentile: using Probability → Value with p = 0.9, the inverse standard normal CDF gives z ≈ 1.2816, so x = 100 + 1.2816×15 ≈ 119.22.

Step 6 — Sanity check with the empirical rule: x = 115 is less than 1 standard deviation above the mean (μ+σ = 115 exactly), consistent with the ~84th percentile falling just past the "68% within ±1σ" mark.

P(X ≤ 115)

≈84.13%

P(X > 115)

≈15.87%

Density at x=115

≈0.016131

90th Percentile Value

≈119.22

Explanation: A value of exactly one standard deviation above the mean (x = μ + σ = 115) lands at approximately the 84.13th percentile — consistent with the empirical rule, since 68% falls within ±1σ, leaving 16% in the upper tail and 84% at or below 115. The PDF value (≈0.016131) is a density, not a probability — it only becomes a probability once integrated (via the CDF) over a range.

Interpretation

Understanding Your Normal Distribution Result

What PDF, CDF, and inverse CDF results each actually mean

Result	What It Means	Example Reading
PDF (density) at x	Relative likelihood near x — a density, not a probability; can exceed 1	0.016131 at x=115 (μ=100, σ=15)
P(X ≤ x) — CDF	The proportion of the distribution at or below x	84.13% of values are ≤ 115
P(X > x) — upper tail	The proportion of the distribution above x; always 1 − CDF	15.87% of values are > 115
Value at percentile p	The x-value below which p×100% of the distribution falls	90th percentile ≈ 119.22

Typical ranges (the empirical rule): about 68% of values fall within μ ± 1σ, 95% within μ ± 2σ, and 99.7% within μ ± 3σ. Values beyond μ ± 3σ are rare enough to often be treated as outliers on a truly normal distribution.

PDF vs CDF/probability: never read a PDF value as a probability by itself — only the area under the PDF curve between two points (given by subtracting two CDF values) represents an actual probability.

Manual verification: confirm P(X ≤ x) + P(X > x) always equals 1, and that a probability of 0.5 in the Probability → Value tab always returns x = μ exactly, since the normal distribution is perfectly symmetric.

Use Cases

Practical Use Cases for the Normal Distribution Calculator

Where modeling a variable as normally distributed is genuinely useful

🎓

Standardized testing cutoffs

Find what percentage of test-takers scored below a given value, or the score cutoff for a target percentile.

🏭

Quality control & Six Sigma

Model manufacturing tolerances and defect rates assuming normally distributed measurement error.

💹

Finance & value-at-risk

Estimate the probability of extreme portfolio moves under a normal returns assumption.

⚕️

Biology & medicine

Model natural variation in height, blood pressure, or other continuous biological measurements.

🔬

Research & confidence intervals

Compute critical values and confidence intervals from the standard normal distribution.

🏗️

Manufacturing tolerance design

Set acceptable measurement ranges based on a target percentage of production falling within spec.

📊

Data science & ML modeling

Model continuous features as normally distributed for statistical tests and anomaly thresholds.

🧬

Genetics & population studies

Model continuous traits like height or weight across a population sample.

🌦️

Weather & climate modeling

Estimate the probability of a temperature or rainfall value falling in a given range.

🏆

Sports performance benchmarking

Estimate what percentile a performance metric falls at relative to a league-wide distribution.

🏥

Insurance & actuarial modeling

Model claim sizes or risk exposure under a normal or near-normal distribution assumption.

🎯

Machine learning anomaly thresholds

Flag data points beyond a chosen percentile or z-score as statistical anomalies.

📈

Grading on a curve

Set grade-boundary percentiles for a class assuming scores are roughly normally distributed.

Pros & Cons

Advantages and Limitations

What this normal distribution calculator does well, and where care is still needed

✅ Advantages

Free, instant, and requires no signup or account
Runs entirely in your browser — no data ever leaves your device
Computes PDF, CDF, and inverse CDF for any mean and standard deviation, not just the standard normal
Reports both the cumulative probability and its complement (upper tail) automatically
Uses well-established polynomial approximations for both the CDF and inverse CDF
Also reports the underlying z-score for cross-checking against a z-table
Useful across education, quality control, finance, and research alike
Validates that the standard deviation is greater than 0 and the probability is strictly between 0 and 1
No calculation limit — use it as many times as needed
Fast-loading and fully mobile-friendly
Removes manual z-table lookup, interpolation, and PDF arithmetic

⚠️ Limitations

The CDF and inverse CDF are numerical approximations, not exact closed-form algebraic results (the normal CDF has no elementary closed form)
Requires the standard deviation to be greater than 0
Rejects a probability of exactly 0 or 1 in the Probability → Value tab, since these correspond to infinite values
Assumes the underlying data is genuinely, or approximately, normally distributed — skewed or bounded real-world data will not match exactly
Does not visualize the bell curve or shade the corresponding area under it
Computes results for one value or probability at a time, not a full table or range
Does not distinguish population parameters from sample estimates — you must supply the correct mean and standard deviation for your context

Reference

PDF vs CDF vs Inverse CDF

Three related but distinct outputs this calculator reports

Concept	Definition	Answers the Question
PDF — Probability Density Function	The relative likelihood (density) of the distribution near a specific value x	"How concentrated is the distribution right at x?"
CDF — Cumulative Distribution Function	The probability that a random value from the distribution is ≤ x	"What share of values fall at or below x?"
Inverse CDF	The value x corresponding to a given cumulative probability p	"What value marks the pth percentile?"

Common Mistakes and Expert Tips

❌ Common Mistakes

Reading a PDF (density) value as if it were a probability — it isn't, and it can exceed 1
Assuming the CDF has an exact algebraic formula rather than a numerical approximation
Applying normal distribution results to data that is actually skewed, bounded, or multimodal
Confusing P(X ≤ x) with P(X = x), which is always 0 for a continuous distribution
Entering a probability of exactly 0 or 1 in the Probability → Value tab and expecting a finite result
Forgetting that the mean and standard deviation must describe the specific population being analyzed

💡 Expert Tips & Best Practices

Use the empirical rule (68-95-99.7) as a fast sanity check before trusting an exact CDF figure
Pair this tool with the Z-Score Calculator when you only need the standardized score, not the full density
Compute σ first with the Standard Deviation Calculator if you're starting from raw data rather than a known parameter
For discrete favorable/total outcome problems instead of continuous data, use the Probability Calculator instead
Always check your data's shape (histogram or normality test) before assuming normality for high-stakes decisions

📝

Summary: This normal distribution calculator gives you an instant, free way to compute the density, cumulative probability, and inverse CDF for any mean and standard deviation — with the formula and a numerically honest note that the CDF has no simple closed form. Pair it with related tools like the Z-Score Calculator and Standard Deviation Calculator for a fuller statistical picture of your data.

FAQ

Frequently Asked Questions

Common questions about the normal distribution

What is the normal distribution?

The normal (or Gaussian) distribution is a symmetric, bell-shaped probability distribution defined by its mean μ (center) and standard deviation σ (spread), used to model many naturally occurring measurements.

What is the difference between PDF and CDF?

The PDF (probability density function) gives the relative likelihood of a value near x, while the CDF (cumulative distribution function) gives the probability that a random value is less than or equal to x — the area under the curve up to x.

How do you find a value from a percentile?

This uses the inverse CDF: first find the z-score corresponding to that percentile on the standard normal distribution, then convert it to your distribution's scale with x = μ + z·σ.

Why does P(X = x) always equal 0 for a continuous distribution?

For continuous distributions like the normal distribution, probability is measured as area under the curve over a range, not at a single point — the area of a single point is 0. That's why we use the PDF for density and the CDF for P(X ≤ x) instead.

What is the empirical rule (68-95-99.7 rule), and how does it apply here?

The empirical rule states that for any normal distribution, about 68% of values fall within 1 standard deviation of the mean, 95% within 2 standard deviations, and 99.7% within 3 standard deviations. You can verify this using the Value → Probability tab: entering x = μ + σ and x = μ − σ and comparing the two cumulative probabilities gives approximately a 68% gap between them.

Does the normal distribution's CDF have an exact closed-form formula?

No. The cumulative distribution function of the normal distribution has no elementary closed-form algebraic expression — it can only be written as an integral of the PDF, or equivalently in terms of the error function (erf). This calculator computes it using the Abramowitz and Stegun polynomial approximation, which is accurate to about 7 decimal places, rather than an exact formula.

Why must the standard deviation be greater than 0?

The PDF, CDF, and z-score formulas all divide by σ, so a value of 0 or a negative number makes every calculation on this page undefined. A standard deviation of 0 would mean every value in the distribution is identical, which isn't a normal distribution at all.

What is the difference between this calculator and the Z-Score Calculator?

The Z-Score Calculator converts a single value to and from a z-score and its percentile. This Normal Distribution Calculator does the same underlying math but also reports the probability density function (PDF) value at a point and lets you work with any mean and standard deviation directly.

Can the normal distribution model real-world data perfectly?

No. The normal distribution is a mathematical idealization. Many real-world datasets — heights, test scores, measurement errors — approximate it closely enough to be useful, but real data often has some skew, outliers, or bounded ranges that a true normal distribution doesn't have. Always check whether your data is reasonably close to normal before relying heavily on these results.

What is the inverse CDF used for?

The inverse CDF (used in the Probability → Value tab) answers "what value corresponds to this percentile?" — for example, finding the 90th-percentile cutoff score on an exam. It works by first finding the z-score for that probability on the standard normal distribution, then converting it to your distribution's scale with x = μ + z·σ.

Why is a probability of exactly 0 or 1 not accepted in the Probability → Value tab?

A cumulative probability of exactly 0 or 1 corresponds to x = −∞ or x = +∞ on a true normal distribution, since the curve's tails never quite reach zero density. This calculator requires a probability strictly between 0 and 1 so it can return a finite, meaningful value.

What does a probability of exactly 0.5 correspond to in the Probability → Value tab?

A probability of 0.5 always corresponds to x = μ, the mean of the distribution, since the normal distribution is symmetric and exactly half its area lies on each side of the mean.

Can I use this calculator for a non-standard normal distribution?

Yes. Unlike a basic z-table, which only covers the standard normal distribution (mean 0, standard deviation 1), this calculator accepts any mean and standard deviation you enter, converting internally to a z-score before applying the standard normal CDF or inverse CDF.

Is the PDF value itself a probability?

No. For a continuous distribution, the PDF gives a density, not a probability — it can even exceed 1 for a very small standard deviation. Only the area under the PDF curve between two points, computed via the CDF, represents an actual probability.

How accurate are the CDF and inverse CDF this calculator reports?

Both use well-established polynomial approximations (Abramowitz and Stegun for the CDF, Beasley-Springer-Moro for the inverse CDF) that are accurate to roughly 6-7 decimal places — more than sufficient for statistics coursework, quality control, and most research applications.

What's the difference between P(X ≤ x) and P(X ≥ x)?

P(X ≤ x) is the CDF value directly. P(X ≥ x), the upper-tail probability, is simply 1 − P(X ≤ x), since the total probability under the curve always sums to 1. This calculator's Value → Probability tab reports both automatically.

Learn More

Authoritative Resources on the Normal Distribution

Trusted educational references to go deeper on the bell curve and its math

🎓 Khan Academy — Statistics & ProbabilityFree lessons and practice on the normal distribution ↗ 🔬 Wolfram MathWorld — Normal DistributionFormal mathematical definition and notation ↗ 📖 OpenStax Introductory Statistics 2eFree, peer-reviewed college statistics textbook ↗ 🏛️ NIST/SEMATECH e-Handbook of Statistical MethodsOfficial U.S. government statistical reference ↗

Related Calculators

Explore other statistics tools