📊 Normal Distribution Calculator – PDF, CDF, Tails, Quantiles & Z-Scores
The normal distribution (also called the Gaussian distribution) is the cornerstone of probability and statistics. Its iconic bell-shaped curve describes how continuous data cluster symmetrically around a central value. This calculator lets you evaluate any aspect of a normal distribution — probability density, cumulative probability, tail probabilities, interval ranges, inverse quantiles, and z-score conversions — for any mean (μ) and standard deviation (σ) you specify.
🔧 Six Calculation Modes
Choose the mode that matches your question:
| Mode | What it answers | Formula |
|---|---|---|
| PDF at x | How likely is the value x (relative density)? | f(x) = (1/σ√2π) · e^(-½·z²) |
| Left-Tail CDF | What fraction of values are ≤ x? | P(X ≤ x) = Φ(z) |
| Right-Tail | What fraction of values are ≥ x? | P(X ≥ x) = 1 − Φ(z) |
| Interval | What fraction of values fall between a and b? | P(a ≤ X ≤ b) = Φ(z₂) − Φ(z₁) |
| Inverse Normal | What x-value corresponds to a given percentile? | x = μ + σ · Φ⁻¹(p) |
| Z-Score | How many SDs away from the mean is x? | z = (x − μ) / σ |
📐 Core Formulas Explained
Every calculation starts by standardizing your value using the z-score formula: z = (x − μ) / σ. This converts any N(μ, σ) problem to the standard normal distribution N(0,1), which has mean 0 and standard deviation 1.
The Probability Density Function (PDF) tells you the relative likelihood of observing a value near x. It does not represent probability directly — since the normal distribution is continuous, the probability at any exact point is zero. PDF is most useful for comparing the relative frequency of different values.
The Cumulative Distribution Function (CDF), written Φ(z), gives the probability that a random observation falls at or below x. The right-tail probability is simply 1 − Φ(z).
📏 The Empirical Rule (68-95-99.7 Rule)
For any normally distributed variable, the following proportions hold regardless of the specific μ and σ:
• 68.27% of values fall within ±1σ of the mean
• 95.45% of values fall within ±2σ of the mean
• 99.73% of values fall within ±3σ of the mean
Use the Interval Probability mode with bounds μ − σ to μ + σ to verify these on your own distribution.
🔄 Inverse Normal and Quantile Lookup
The Inverse Normal mode is the reverse of the CDF: you provide a target probability p (e.g., 0.95) and the calculator finds the raw-score value x such that P(X ≤ x) = p. This is commonly used to find:
• Percentile cutoffs — e.g., the 90th-percentile exam score
• Critical values — z = 1.96 for a two-tailed 95% test
• Confidence interval bounds in statistics coursework
• Process tolerance limits in quality control
🏭 Practical Applications
Normal distribution calculations appear in diverse real-world contexts:
• Education – grading on a curve, test-score analysis, admissions cutoffs
• Quality control – process capability (Cp, Cpk), defect-rate estimation
• Finance – returns modeling, Value-at-Risk, option pricing (Black-Scholes)
• Medicine – reference ranges for lab values, clinical trial endpoints
• Engineering – tolerance analysis, reliability estimation, signal noise
📊 Standard Normal Shortcut
Enable the Use Standard Normal toggle to lock μ = 0 and σ = 1. In this mode, your x-value is interpreted directly as a z-score, and all outputs map to the standard N(0,1) distribution. This is equivalent to looking up a z-table but with instant, high-precision results for any probability you need.
⚠️ When the Normal Model May Not Apply
The normal distribution assumes symmetric, unimodal data with no heavy tails. It may be a poor fit for:
• Skewed data — income, survival times, reaction rates
• Bounded variables — percentages, counts, ratings
• Heavy-tailed distributions — financial returns, earthquake magnitudes
• Small samples — use the t-distribution instead
Always verify normality with a histogram, Q-Q plot, or a formal test (Shapiro-Wilk, Anderson-Darling) before relying on normal-distribution probabilities.