Central Limit Theorem

As explained in Experimentation for Engineers

A theorem stating that suitably normalized sums or averages of many observations tend toward a normal distribution.

The Central Limit Theorem makes a normal distribution a practical approximation for standardized aggregate measurements as the number of individual measurements increases. It supports engineering decisions while preserving uncertainty: the approximation is useful, not exact. section 2.3.1

A/B testing with z scores

Compare execution costs on ASDAQ and BYSE, compute the difference and its standard error, then standardize the result as z. Under the unit-normal approximation, z < -1.64 marks the lower 5% tail: act as though BYSE is better. The cutoff controls false positives at 5%, but the decision remains a bet rather than a certainty. section 2.3.1

Figure 2.8 The distribution of z scores has mean zero (by assumption) and standard deviation one. The shaded area to the left of the vertical line is 5% of the total shaded area. The probability of a measurement of z having a value to the left of the vertical line is 5%.

Use the unit-normal histogram to inspect the 5% tail and the z = -1.64 decision boundary. section 2.3.1

How Experimentation for Engineers covers it

How deep it goes

Explains the increasing-sample approximation and applies a unit-normal cutoff, but does not derive convergence or assumptions in detail.

Examples

Uses ASDAQ-versus-BYSE execution costs, a z = -1.64 cutoff, and a unit-normal tail visualization.

What it assumes

Assumes aggregate measurements, standard errors, z scores, and normal-distribution tail probabilities.

What you can do after

Evaluate A/B-test evidence with a normal approximation while recognizing false-positive risk and business trade-offs.

How it connects

Connects aggregate measurements, standard errors, z scores, normal distributions, and false positives in A/B testing.

liveBookmanning.com

Read Experimentation for Engineers on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Grokking StatisticsA theorem stating that, under suitable conditions, the distribution of sample means approaches a normal distribution as sample size increases, regardless of the population distribution's shape.
• 

  liveBookmanning.com
  Causal Inference for Data ScienceThe central limit theorem states that, under suitable independence and finite-variance conditions, a normalized sample mean approaches a normal distribution as sample size grows.

Compare how 3 books explain Central Limit Theorem →

Related concepts

• normal distributionproduces · A normal distribution is a continuous probability distribution with a symmetric, bell-shaped density determined by its mean and variance.
• sampling distributionproduces · A sampling distribution is the probability distribution of a statistic across repeated samples drawn from a specified population or process.
• meanuses · A mean is a measure of central tendency, calculated as the arithmetic average of values or, for a random variable, as the probability-weighted average of its possible values.
• student's t-distributioncontrasts with · Student’s t-distribution is a family of continuous probability distributions with heavier tails than the normal distribution, used in inference when variability is estimated from data.

This explainer was generated by Manning's knowledge graph from the text of Experimentation for Engineers. The section links point to the parts of the book it draws on.