Inverse Normal and Parameters (AI HL)

An inverse normal calculation finds a value xx from a given probability, where the probability is the area to the left of xx. When the mean or standard deviation is unknown, the zz-value z=x−μσz=\dfrac{x-\mu}{\sigma} turns the probability statement into an equation to solve. The zz-value comes from the GDC, and the equation is solved by rearranging. It's part of the broader Probability & Distributions topic.

24 questions on this sub-topic.

Syllabus

CodeSyllabus content
SL4.9The normal distribution and curve. Properties of the normal distribution, including that approximately 68% of the data lies between μ±σ\mu\pm\sigma, 95% between μ±2σ\mu\pm2\sigma and 99.7% between μ±3σ\mu\pm3\sigma. Normal probability calculations and inverse normal calculations must be found using technology.

Official IB source: Mathematics: applications and interpretation subject brief (PDF, ibo.org).

Skills practised on this page:

Standardising and going backwards

Neither relationship below is in the formula booklet - you're expected to know the standardising equation and to drive the GDC's inverse normal routine confidently.

Standardising a value

z=x−μσz = \dfrac{x-\mu}{\sigma}

Converts any normal variable X∼N(μ,σ2)X\sim N(\mu,\sigma^2) to the standard normal zz-score. Useful when a parameter (μ\mu or σ\sigma) is the unknown, since it turns the problem into an equation.

Not in the formula booklet

Inverse normal

Given P(X<x)=pP(X<x)=p, the GDC's inverse normal routine returns xx directly.

Always enter the area to the left of the unknown value - if you're told "the top k%k\%", convert it to 1−k1001-\tfrac{k}{100} first.

Not in the formula booklet

Need the full syllabus wording and formula-booklet reference table? See Probability & Distributions. GDC keystrokes for these problems are in the same page's GDC section.

Key terms

What is a zz-value?

A zz-value is the number of standard deviations a value lies above the mean, given by z=x−μσz=\dfrac{x-\mu}{\sigma}. It is negative for values below the mean.

e.g. For X∼N(40,52)X\sim N(40,5^2), the value x=50x=50 has z=50−405=2z=\dfrac{50-40}{5}=2.

How do you find a boundary value from an area to the left?

Enter the area to the left of the boundary into the inverse normal function with the mean and standard deviation. The output is the boundary value.

e.g. For the standard normal Z∼N(0,1)Z\sim N(0,1), P(Z<a)=0.975P(Z<a)=0.975 gives a≈1.96a\approx1.96.

How do you find an unknown mean or standard deviation?

Find the zz-value for the given probability using inverse normal on N(0,1)N(0,1), then solve z=x−μσz=\dfrac{x-\mu}{\sigma} for the unknown.

e.g. If X∼N(μ,22)X\sim N(\mu,2^2) and P(X<18)=0.975P(X<18)=0.975, then 18−μ2=1.96\dfrac{18-\mu}{2}=1.96, so μ≈14.1\mu\approx14.1.

How does symmetry help with inverse normal?

Boundaries with equal areas in the two tails lie the same distance either side of the mean. So the upper boundary can be found from the lower one by reflecting in the mean.

e.g. For X∼N(60,102)X\sim N(60,10^2), the value with 10% below it is about 47.247.2, so the value with 10% above it is about 72.872.8.

Worked examples

1
Medium
GDC
[4 marks]

Lifetimes of a battery are normally distributed with mean 80 hours and standard deviation 8 hours.

(a) Find the lifetime exceeded by 90% of batteries.

(b) The manufacturer guarantees a minimum lifetime so that only 5% fail early.

Find that minimum lifetime.

Worked solution

(a) P(X>L)=0.90⇒P(X<L)=0.10P(X > L) = 0.90 \Rightarrow P(X < L) = 0.10 M1
L≈69.7L \approx 69.7 h. A1

(b) P(X<m)=0.05P(X < m) = 0.05 M1
m≈66.8m \approx 66.8 h. A1
m≈66.8m \approx 66.8 h. On the GDC. (a) Find kk such that P(X<k)=0.1P(X<k)=0.1, where X∼N(80, 82)X\sim N(80,\ 8^2). (b) Find kk such that P(X<k)=0.05P(X<k)=0.05, where X∼N(80, 82)X\sim N(80,\ 8^2).

TI-84 Plus CE: (a) 2nd vars 3:invNorm( area 0.1 mean 80 sd 8 enter Then (b) 2nd vars 3:invNorm( area 0.05 mean 80 sd 8 enterCasio fx-CG50: (a) MENU Statistics F5 F1 F3 F2 ▼ ▼ Area 0.1 EXE σ 8 EXE μ 80 EXE ▼ F1 Then (b) MENU Statistics F5 F1 F3 F2 ▼ ▼ Area 0.05 EXE σ 8 EXE μ 80 EXE ▼ F1Casio fx-CG100: (a) HOME Distribution Normal OK Tail X ≤ (left tail) any x, then μ 80 OK σ 8 OK Execute OK select p, type 0.1 OK Then (b) HOME Distribution Normal OK Tail X ≤ (left tail) any x, then μ 80 OK σ 8 OK Execute OK select p, type 0.05 OKTI-Nspire CX: (a) menu Probability Distributions Inverse Normal Area 0.1 tab μ 80 tab σ 8 enter Then (b) menu Probability Distributions Inverse Normal Area 0.05 tab μ 80 tab σ 8 enterFull GDC guide: Inverse normal Topic page

M1 Inverse normal A1 \approx69.7 \\approx69.7\ M1 InvNorm 0.05 A1 \approx66.8 \\approx66.8\
2
Medium
Calculator
[4 marks]

IQ scores are normal with mean 100 and standard deviation 15.

(a) Find the upper quartile.

(b) Find the interquartile range.

Worked solution

(a) P(X<Q3)=0.75P(X < Q_3) = 0.75 M1
Q3≈110.1.Q_3 \approx 110.1. A1

(b) By symmetry M1
Q1≈89.9Q_1 \approx 89.9 A1
IQR ≈20.2.\approx 20.2.

M1 Inverse normal A1 \approx110.1 \\approx110.1\ M1 \Q1 Q_1\ A1 \89.9 89.9\
3
Easy
GDC
[2 marks]

Test scores are N(60,102).N(60, 10^{2}). Find the probability a score is above 70

Worked solution

Scores X∼N(60,102).X\sim N(60,10^2). ‘Above 70’ is a right tail. M1
7070 is one standard deviation above the mean - the familiar ≈16%\approx 16\%, giving 0.159.0.159. A1
On the GDC. Find P(X>70)P(X>70) for X∼N(60, 102)X\sim N(60,\ 10^2). The calculator shows about 0.159.

TI-84 Plus CE: 2nd vars 2:normalcdf( lower 70 upper 1 2nd , 9 9 mean 60 sd 10 enterCasio fx-CG50: MENU Statistics F5 F1 F2 F2 ▼ Lower 70 EXE Upper 1 EXP 9 9 EXE σ 10 EXE μ 60 EXE ▼ F1Casio fx-CG100: HOME Distribution Normal OK Tail X ≥ (right tail) Lower 70 OK μ 60 OK σ 10 OK Execute OKTI-Nspire CX: menu Probability Distributions Normal Cdf Lower 70 tab Upper 1E99 tab μ 60 tab σ 10 enterFull GDC guide: Normal distribution probabilities Topic page

M1 Right tail / normalcdf set-up A1 Correct answer of 0.1590.159
4
Hard
GDC
[4 marks]

X∼N(100,102).X\sim N(100,10^2). For a sample of 4, find P(X¯>108).P(\bar X>108).

Worked solution

X¯∼N(100,25)\bar X\sim N(100,25), standard deviation 5. M1
P(X¯>108)P(\bar X>108) by technology M1
P(X¯>108)≈0.0548.P(\bar X>108)\approx0.0548. A1
≈0.0548.\approx0.0548. A1On the GDC. Find P(X>108)P(X>108) for X∼N(100, 52)X\sim N(100,\ 5^2). The calculator shows about 0.0548.

TI-84 Plus CE: 2nd vars 2:normalcdf( lower 108 upper 1 2nd , 9 9 mean 100 sd 5 enterCasio fx-CG50: MENU Statistics F5 F1 F2 F2 ▼ Lower 108 EXE Upper 1 EXP 9 9 EXE σ 5 EXE μ 100 EXE ▼ F1Casio fx-CG100: HOME Distribution Normal OK Tail X ≥ (right tail) Lower 108 OK μ 100 OK σ 5 OK Execute OKTI-Nspire CX: menu Probability Distributions Normal Cdf Lower 108 tab Upper 1E99 tab μ 100 tab σ 5 enterFull GDC guide: Normal distribution probabilities Topic page

M1 Distribution of X¯\bar X M1 Normalcdf A1 Correct answer of 0.0548 A1 3 significant figures

Common mistakes

  • Forgetting inverse normal needs the area to the LEFT. For "the top 10%", the area to enter is 0.900.90, not 0.100.10 - misreading this flips the answer to the wrong tail entirely.
  • Not identifying which quantity is actually unknown. Sometimes it's xx you're solving for, sometimes it's μ\mu or σ\sigma - each needs the standardising equation rearranged differently, and treating every question as "find xx" leads to the wrong setup.
  • Rounding a zz-value or intermediate probability too early. Carrying a rounded zz-score into a further calculation for μ\mu or σ\sigma can shift the final answer outside the accuracy the mark scheme accepts - keep full GDC precision until the last line.

Using your GDC

Every step below is a real button sequence, not a vague "use your calculator" hint - covering the TI-84 Plus, TI-Nspire, and Casio fx-9860/fx-CG50. Pick your model to filter down to just the steps that apply to you.

Show steps for:
Inverse normal (find the value for a given probability)

The reverse of a normal probability: given a percentage, find the cut-off value (e.g. the mark for the top 10%).

  1. Work out the area to the LEFT of the value you want.
  2. 2nd → VARS (DISTR) → invNorm(area, μ, σ). Newer OS lets you pick the tail.TI-84
  3. menu → Probability → Distributions → Inverse Normal; enter the area, μ and σ.Nspire
  4. Main menu → Statistics → DIST → NORM → InvN; set the tail and enter area, σ, μ.Casio

Tip: invNorm needs the area to the LEFT. For "top 10%", use area = 0.90; for "bottom 25%", use area = 0.25.

Find an unknown mean or standard deviation (normal)

Given a probability and one parameter, work back to the missing μ\mu or σ\sigma - a standard exam twist.

  1. Turn the probability into a zz-value using the inverse normal with μ=0\mu = 0, σ=1\sigma = 1.
  2. invNorm(area-to-left, 0, 1) gives zz; then solve z=(x−μ)/σz = (x-\mu)/\sigma for the unknown.TI-84
  3. Use Inverse Normal with μ=0\mu=0, σ=1\sigma=1 to get zz, then solve the standardising equation for μ\mu or σ\sigma.Nspire
  4. DIST → NORM → InvN with μ=0\mu=0, σ=1\sigma=1 gives zz; substitute into z=(x−μ)/σz=(x-\mu)/\sigma.Casio
  5. If two probabilities are given, form two equations and solve them simultaneously for μ\mu and σ\sigma.

Tip: Always sketch the curve first so you can tell which side of the mean each value should fall on.

See the full GDC guide for more calculator models and topics.

Quick questions

What does "inverse normal" mean?

A normal probability calculation works forwards: given xx, find P(X<x)P(X<x). An inverse normal calculation works backwards: given a probability (an area under the curve), find the value of xx that produces it.

Which area do I enter for "the top 10%"?

Inverse normal functions on the GDC always take the area to the left of the unknown value. For "the top 10%", the boundary has 90% of the distribution below it, so you enter 0.900.90, not 0.100.10.

Why is my inverse normal answer a negative zz-value?

A negative zz-value means the boundary is below the mean, which happens when the area to the left is less than 0.50.5. For example, an area of 0.20.2 to the left gives z≈−0.842z\approx-0.842.

What if both the mean and the standard deviation are unknown?

Two probability statements give two equations in μ\mu and σ\sigma, which are solved simultaneously. For example, P(X<20)=0.1P(X<20)=0.1 and P(X<40)=0.9P(X<40)=0.9 give z=∓1.2816z=\mp1.2816, so by symmetry μ=30\mu=30 and σ=101.2816≈7.80\sigma=\dfrac{10}{1.2816}\approx7.80.

How can I check an inverse normal answer is sensible?

Sketch the curve and compare the answer with the mean. If the area to the left is more than 0.50.5, the boundary must be above the mean, and if it is less than 0.50.5, the boundary must be below it.

Ready to practise properly?

24 inverse-normal questions, marked instantly like the real exam.