MGF of the Normal Explained with Worked Examples

For the moment generating function is .

Write 2x or 2*x to multiply, ^ for powers, and ; between simultaneous equations. You can also type factor, expand, derivative of, integrate … from a to b, limit … as x->0 or 15% of 80.

Try an example

Before you start

This page is easier if you are comfortable with:

What MGF of the normal means

It follows by completing the square in the exponent of (completing the square, the Gaussian integral). For , : , so , (the mean) and , giving variance 4. The density is in the normal density.

The exponent is quadratic in , and sums of independent normals add: has MGF , i.e. ; at it equals . This reproduces combining normal variables and sums of independent variables.

Because only the first two cumulants are non-zero, the MGF characterises the normal distribution. It is the limit of standardised MGFs in the proof of the central limit theorem, and the bound gives tail bounds (Chernoff and Hoeffding bounds). General theory: moment generating functions.

What is it used for?

The normal MGF has a quadratic exponent, which is why sums of normals are normal.

How to use mgf of the normal

  1. Identify and .
  2. Write .
  3. Differentiate at 0 for moments.
  4. For sums of independent normals add means and variances.

Graph

MGF exp(t + 2t²) of a normal variable with mean 1 and variance 4t = 0.5: 2.7183-1.4-1.2-1-0.8-0.6-0.4-0.200.20.40.60.81051015202530MGF exp(t + 2t²) of a normal variable with mean 1 and variance 4tM(t)M(t)Graph with x from -1.5 to 1 and y from 0 to 30. Shows the curve M(t); point t = 0.5: 2.7183 at (0.5, 2.7183).

The MGF grows very fast because of the t squared term.

Worked examples

These examples are worked by hand, step by step, so you can compare your own working.

Example 1: M(0.5)

Find for .

Worked example
Answer
2.71832.7183
  1. Exponent

Example 2: second moment

Find for .

Worked example
Answer
55
  1. M″(0)

Example 3: a sum of normals

Find the MGF at of the sum of independent and .

Worked example
Answer
1808.041808.04
  1. Sum is N(3, 9)

Special cases

A few situations come up again and again. Knowing them saves time and helps you spot answers that cannot be right.

N(1, 4), t = 0.5

exp(0.5 + 0.5)

2.7183.

Mean

M′(0) = μ

1.

E(X²)

μ² + σ²

5.

Independent sum

N(1, 4) + N(2, 9)

N(3, 13).

Word problems

For X ~ N(μ, σ²), M(t) = exp(μt + σ²t²/2). So M′(0) = μ, M″(0) = μ² + σ², and independent normals add means and variances.

Evaluate

Find M(0.5) for N(1, 4).

  1. exp(0.5 + 0.5)
  2. Check: ✓2.7183

Mean

Find M′(0).

  1. μ
  2. Check: ✓1

Second moment Higher

Find E(X²).

  1. 1 + 4
  2. Check: ✓5

Sum Higher

Find the variance of N(1, 4) + N(2, 9).

  1. 4 + 9
  2. Check: ✓13

Common mistakes

σ instead of σ²

σt²/2 ✗

σ²t²/2 ✓

Variance, not sd.

t² missing

σ²t/2 ✗

σ²t²/2 ✓

Square t.

Adding sds

2 + 3 ✗

add variances ✓

Not standard deviations.

M(0)

not 1 ✗

M(0) = 1 ✓

Check.

Exam tips

  1. exp(μt + σ²t²/2).
  2. M′(0) = μ.
  3. M″(0) = μ² + σ².
  4. Add variances.
  5. Standard normal: exp(t²/2).

Practice questions

Try each one on paper, then open the answer. Use the solver above to see the full working.

1. M(1) for N(0, 1) to 4 d.p. Foundation

1.6487

2. M(2) for N(0, 1) to 4 d.p. Foundation

7.3891

3. M(1) for N(2, 4) to 4 d.p. Foundation

54.5982

4. E(X²) for N(3, 4) Higher

13

5. Variance of N(0,1)+N(0,3) Higher

4

6. M(0.5) for N(0, 4) to 4 d.p. Higher

1.6487

Practice generator

Unlimited fresh questions with instant marking. Choose a level, work it out on paper, then type your answer.

Solve: …

Key terms

Normal MGF
exp(μt + σ²t²/2).
Mean (μ)
Centre.
Variance (σ²)
Spread squared.
Standard normal
μ = 0, σ = 1.
Completing the square
Derivation step.
Independent sum
Normal again.

Frequently asked questions

What is the normal MGF?

exp(μt + σ²t²/2).

What is E(X²)?

μ² + σ².

What is the sum of independent normals?

Normal with summed mean and variance.

What is the standard normal MGF?

exp(t²/2).

Which page covers the Poisson MGF?

MGF of the Poisson.

Which page covers Chebyshev?

Chebyshev’s inequality.

More in Advanced statistics and stochastic processes and the formula sheets.