Probability Lab

Stochastic tables, Bayes’ theorem, eikosograms

Size

Problem

Show probabilities as

Mode

Typing answers and formulas

Probabilities can be typed as fractions (2/5), decimals (0.4) or percentages (40%). Fractions don’t need to be reduced: 4/10 counts as 2/5.

¬: ~ !
∧: & ^
∨: | v (between spaces)
→: -> =>
↔: <-> <=>

0 / 0

Sentence letters

Problem

Show probabilities as

Mode

The method

As in the book: start from the dictionary, decide whether you learned E or ¬E, identify the prior and the likelihoods, then use Bayes’ theorem to compute the posterior.

The likelihoods are always about E: P(E | H) and P(E | ¬H). If you learned ¬E, use P(¬E | H) = 1 − P(E | H) and P(¬E | ¬H) = 1 − P(E | ¬H).

Probabilities can be typed as fractions (7/26), decimals (0.27) or percentages (27%).

0 / 0

1. What did you learn?

2. Prior and likelihoods

P(H) = P(E | H) = P(E | ¬H) =

3. Posterior

P(H | E) =

Eikosogram Studio

Joint probability, drawn to scale

Text size

Probabilities

P(H)

Width of the left column

P(E|H)

Height of the bottom cell, left column

P(E|not H)

Height of the bottom cell, right column

Labels

Title

Left column

Right column

Bottom row

Top row

Export for slides

Transparent background

SVG stays sharp at any size in Keynote and PowerPoint. The exported figure is always light, and includes the P(H|E) bar when it’s showing.

Prior50.0%P(H)

Evidence50.0%P(E)

Posterior50.0%P(H|E)

If not E50.0%P(H|not E)

Probability Eikosogram

Show P(H|E)

Drag a divider or handle to reshape the diagram. On a keyboard, Tab to a divider and use the arrow keys (hold Shift for bigger steps).