ECON4916 Chap.1 Probability Tools for Financial Data
Probability Tools for Financial Data
Financial econometrics begins by defining what is random and how possible values receive probability. This chapter separates a random variable from its realised observation, maps discrete support to probability masses, interprets continuous probability through areas under a density, and uses the cumulative distribution to answer threshold questions.
It then builds means, variances and covariances from their definitions and explains why moment existence matters for heavy-tailed financial quantities. The final section distinguishes independence from zero covariance with a nonlinear counterexample.
The practical aim is to make every probability calculation auditable: state the support, choose a sum or integral, verify normalisation, preserve units and limit the interpretation to what the computed object actually measures.
What this chapter covers
- 01
Sample spaces and random variables
- 02
Discrete probability mass functions
- 03
Continuous densities and cumulative distributions
- 04
Expected values and raw moments
- 05
Variance and covariance
- 06
Independence and nonlinear dependence
Worked example · free
Moments of a three-outcome return
- +1Add the masses: 0.20 + 0.50 + 0.30 = 1.00, and confirm that all are nonnegative.
- +1Compute E[X] = (-2)(0.20) + 0(0.50) + 3(0.30) = 0.50 percentage points.
- +1Compute E[X squared] = 4(0.20) + 0 + 9(0.30) = 3.50 squared percentage points.
- +1Use Var(X) = E[X squared] - E[X] squared = 3.50 - 0.25 = 3.25.
- +1Interpret 0.50 as a probability-weighted centre across repeated draws and 3.25 as squared dispersion, not a guaranteed outcome.
Key terms
- Sample Space
- The complete set of possible outcomes underlying a probability experiment or stochastic model.
- Random Variable
- A function mapping each outcome in a sample space to a real number.
- Probability Mass
- The probability assigned to one support point of a discrete random variable.
- Probability Density
- A continuous-distribution function whose integral over an interval equals the interval probability.
- Cumulative Distribution
- The probability that a random variable is less than or equal to a specified threshold.
- Second Moment
- The expected squared value of a random variable, used with the mean to calculate variance.
Probability Tools for Financial Data FAQ
Why can a density be greater than one?
A density height is not a probability. Probability is the integrated area over an interval, so a narrow distribution may need a height above one while its total area remains exactly one. For a continuous variable, any exact point still has probability zero because that point has zero width.
When should I use a sum instead of an integral?
Use a sum when the random variable is discrete and probability is attached to separate support points. Use an integral when the variable is continuous and described by a density over intervals. Writing the support first usually makes the correct operator obvious and prevents accidental calculation outside the possible values.
Does zero covariance imply independence?
No. Zero covariance removes linear association, but a nonlinear relationship can remain. If Y is the square of a symmetric variable X, covariance may be zero even though Y is completely determined by X. Independence must be checked through the joint distribution or by comparing conditional and marginal probabilities.
Why check whether a moment exists?
A formula for variance is meaningful only when the second moment is finite. Heavy-tailed distributions can have a defined centre or median while lacking finite variance. State the existence condition before using a moment because otherwise a familiar calculation may be manipulating an object the model does not possess.
Exam move
Build probability fluency through representation changes. For each practice variable, write its sample space, support and distribution before calculating anything. Translate a discrete table into a cumulative distribution and check that the final cumulative value is one. For continuous cases, sketch the support, mark the requested interval and treat probability as area rather than height.
Recalculate the same mean and variance from definitions instead of relying on a memorised shortcut, then label the units. Keep a separate page of counterexamples: a density above one, a heavy-tailed variable without finite variance, and dependent variables with zero covariance. These examples are more durable than verbal warnings because they expose precisely which implication fails.
Finish each question with a sentence scoped to the object calculated. A mean is a repeated-draw centre, variance is dispersion in squared units, covariance is linear comovement, and independence is a full-distribution claim. When using MATLAB, verify that the support, probability totals and hand-computed moments match the software output before moving to a larger dataset.
Working through Probability Tools for Financial Data in ECON4916? Sia is AskSia’s AI Economics tutor — ask any ECON4916 Probability Tools for Financial Data question and get a clear, step-by-step explanation grounded in how ECON4916 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.