Variance: The average of the squared differences from the mean.
Formula: Var(X) = E[(X - μ)²] = E[X²] - (E[X])²
Standard Deviation: σ = √Var(X), measures spread in original units.
- Calculate E[X] (expected value)
- Calculate E[X²] (expected value of squares)
- Use formula: Var(X) = E[X²] - (E[X])²
- Calculate SD(X) = √Var(X)
E[X] = 1/6 + 2/6 + 3/6 + 4/6 + 5/6 + 6/6 = 21/6 = 3.5
E[X²] = 1/6 + 4/6 + 9/6 + 16/6 + 25/6 + 36/6
= (1 + 4 + 9 + 16 + 25 + 36)/6 = 91/6 ≈ 15.167
Var(X) = E[X²] - (E[X])²
= 91/6 - (21/6)²
= 91/6 - 441/36
= 546/36 - 441/36 = 105/36 = 35/12 ≈ 2.917
SD(X) = √Var(X) = √(35/12) ≈ 1.708
The variance of rolling a fair six-sided die is 35/12 ≈ 2.917.
The standard deviation is √(35/12) ≈ 1.708.
• Variance Formula: Var(X) = E[X²] - (E[X])²
• Standard Deviation: SD(X) = √Var(X)
• Units: Variance in squared units, SD in original units
Risk: Measured by variance/standard deviation - higher values indicate greater uncertainty.
Investment Risk: Higher standard deviation means more volatility in returns.
E[X] = 1.5% + 4.0% + 4.5% = 10%
E[X²] = 7.5%² + 40.0%² + 67.5%² = 115%²
Var(X) = E[X²] - (E[X])² = 115%² - (10%)² = 115%² - 100%² = 15%²
SD(X) = √15%² ≈ 3.87%
E[Y] = 0.4% + 4.8% + 4.0% = 9.2%
E[Y²] = 0.8%² + 38.4%² + 80.0%² = 119.2%²
Var(Y) = E[Y²] - (E[Y])² = 119.2%² - (9.2%)² = 119.2%² - 84.64%² = 34.56%²
SD(Y) = √34.56%² ≈ 5.88%
Investment A: E[X] = 10%, SD(X) = 3.87%
Investment B: E[Y] = 9.2%, SD(Y) = 5.88%
Investment B has higher risk (higher standard deviation) despite slightly lower expected return.
Investment A: Variance = 15%², Standard Deviation ≈ 3.87%
Investment B: Variance = 34.56%², Standard Deviation ≈ 5.88%
Investment B is riskier due to higher standard deviation.
• Variance Formula: Var(X) = E[X²] - (E[X])²
• Risk Assessment: Higher SD indicates higher risk
• Trade-off: Higher expected return may come with higher risk
Winnings Distribution: P(X = $4) = 0.5, P(X = -$2) = 0.5.
Spread Measurement: Variance quantifies how far values deviate from the mean.
E[X] = $2 + (-$1) = $1
E[X²] = $8 + $2 = $10
Var(X) = E[X²] - (E[X])²
= $10 - ($1)²
= $10 - $1 = $9
SD(X) = √$9 = $3
Expected winnings per flip: $1
Typical deviation from expected value: $3
Large standard deviation indicates high variability in outcomes.
The variance of winnings per flip is $9.
The standard deviation is $3, indicating significant variability in outcomes.
• Variance Formula: Var(X) = E[X²] - (E[X])²
• Interpretation: SD shows typical deviation from mean
• Units: Variance in squared units, SD in original units
Variance: The average of the squared differences from the mean, measuring spread of the distribution.
Standard Deviation: The square root of variance, expressed in the same units as the original data.
Dispersion: How spread out the values of a random variable are around the mean.
Units: Variance in squared units, standard deviation in original units.
• Variance: Var(X) = E[X²] - (E[X])²
• Standard Deviation: SD(X) = √Var(X)
• Linear Transformation: Var(aX + b) = a²Var(X)
• Constant: Var(c) = 0
Linear Transformation: When Y = aX + b, the mean and variance transform predictably.
Scaling Effect: Variance is affected by the square of the scaling factor.
E[X] = μ = 5
Var(X) = σ² = 4
Y = 3X - 2 (so a = 3, b = -2)
Using the linear property: E[aX + b] = aE[X] + b
E[Y] = E[3X - 2] = 3E[X] - 2 = 3(5) - 2 = 15 - 2 = 13
Using the variance transformation property: Var(aX + b) = a²Var(X)
Var(Y) = Var(3X - 2) = 3²Var(X) = 9 × 4 = 36
SD(Y) = √Var(Y) = √36 = 6
Var(Y) = E[Y²] - (E[Y])²
Y² = (3X - 2)² = 9X² - 12X + 4
E[Y²] = E[9X² - 12X + 4] = 9E[X²] - 12E[X] + 4
Since Var(X) = E[X²] - (E[X])², we have E[X²] = Var(X) + (E[X])² = 4 + 25 = 29
So E[Y²] = 9(29) - 12(5) + 4 = 261 - 60 + 4 = 205
Var(Y) = 205 - 13² = 205 - 169 = 36 ✓
For Y = 3X - 2: E[Y] = 13, Var(Y) = 36, SD(Y) = 6.
The mean was shifted by -2 and scaled by 3, while variance was scaled by 3² = 9.
• Mean Transformation: E[aX + b] = aE[X] + b
• Variance Transformation: Var(aX + b) = a²Var(X)
• Standard Deviation: SD(aX + b) = |a|SD(X)
Probability Distribution: Complete specification of all possible values and their probabilities.
Variance Calculation: Requires computing E[X] and E[X²] from the distribution.
E[X] = 0×0.1 + 1×0.3 + 2×0.4 + 3×0.2
= 0 + 0.3 + 0.8 + 0.6 = 1.7
E[X²] = 0²×0.1 + 1²×0.3 + 2²×0.4 + 3²×0.2
= 0 + 0.3 + 1.6 + 1.8 = 3.7
Var(X) = E[X²] - (E[X])²
= 3.7 - (1.7)²
= 3.7 - 2.89 = 0.81
SD(X) = √Var(X) = √0.81 = 0.9
Sum of probabilities: 0.1 + 0.3 + 0.4 + 0.2 = 1.0 ✓
The variance of X is 0.81.
The standard deviation of X is 0.9.
• Variance Formula: Var(X) = E[X²] - (E[X])²
• Standard Deviation: SD(X) = √Var(X)
• Verification: Probabilities must sum to 1
Complete Summary: Variance and Standard Deviation of Random Variables
Variance: The average squared deviation from the mean, measuring spread.
Formula: Var(X) = E[(X - μ)²] = E[X²] - (E[X])².
Standard Deviation: σ = √Var(X), in same units as original variable.
Interpretation: Higher values indicate greater dispersion of values.
Non-negativity: Var(X) ≥ 0, with equality only for constants.
Linear Transformation: Var(aX + b) = a²Var(X).
Standard Deviation: SD(aX + b) = |a|SD(X).
Units: Variance in squared units, SD in original units.
Step 1: Calculate E[X] (the mean).
Step 2: Calculate E[X²] (expected value of squares).
Step 3: Apply variance formula: Var(X) = E[X²] - (E[X])².
Step 4: Calculate SD(X) = √Var(X).
Organize Calculations: Use tables to track x, P(X=x), x×P, x²×P.
Check Work: Verify probabilities sum to 1.
Units Matter: Variance has squared units, SD has original units.
Interpretation: SD tells you typical deviation from the mean.
Variance: A measure of how far a set of numbers is spread out from their average value. It's the expectation of the squared deviation of a random variable from its mean.
Standard Deviation: The square root of variance, providing a measure of dispersion that is in the same units as the original data. It represents the average distance of data points from the mean.
Dispersion: The degree to which data is spread out or scattered around the central tendency.
Deviation: The difference between an observed value and the expected value (mean).
Squared Deviation: The square of the deviation, ensuring all values are positive and emphasizing larger deviations.
- Verify Distribution: Ensure all probabilities sum to 1.
- Calculate Mean: E[X] = Σ [x × P(X = x)].
- Calculate E[X²]: Σ [x² × P(X = x)].
- Apply Formula: Var(X) = E[X²] - (E[X])².
- Calculate SD: SD(X) = √Var(X).
- Interpret Results: Explain in context of the problem.
• Variance Formula: Var(X) = E[X²] - (E[X])²
• Standard Deviation: SD(X) = √Var(X)
• Linear Transformation: Var(aX + b) = a²Var(X)
• Constant: Var(c) = 0
Questions & Answers
Question: Why do we square the deviations when calculating variance? Why not just use absolute values?
Answer: Great question! Squaring deviations serves several important purposes:
Mathematical Advantages:
- Squaring ensures all deviations are positive (no cancellation)
- Creates a smooth, differentiable function (important for calculus-based statistics)
- Emphasizes larger deviations more than smaller ones (quadratic effect)
Why not absolute values? We could use absolute deviations (mean absolute deviation), but variance has superior mathematical properties:
- Variance leads to the normal distribution in the Central Limit Theorem
- More amenable to mathematical analysis and optimization
- Preserves important relationships like Var(X+Y) = Var(X) + Var(Y) for independent variables
The standard deviation (square root of variance) brings us back to original units while preserving these mathematical advantages.
Question: In Exercise 2, Investment B had a lower expected return but higher standard deviation. What does this mean for an investor?
Answer: This represents a classic risk-return tradeoff:
Investment A: Higher expected return (10% vs 9.2%) with lower risk (SD = 3.87%)
Investment B: Lower expected return but potentially higher upside with higher risk (SD = 5.88%)
Investor Implications:
- Risk-Averse Investor: Would prefer Investment A (higher return with less risk)
- Risk-Tolerant Investor: Might accept Investment B for potential higher returns
- Efficient Market Theory: Generally, higher risk should correspond to higher expected return
In this case, Investment B offers less reward per unit of risk (lower Sharpe ratio), making Investment A more attractive from a risk-adjusted perspective.
Question: Why does Var(aX + b) = a²Var(X)? Why is the constant b eliminated and why is a squared?
Answer: Let's prove this property step by step:
Definition: Var(Y) = E[(Y - E[Y])²]
Let Y = aX + b, then:
E[Y] = E[aX + b] = aE[X] + b
Now, Y - E[Y] = (aX + b) - (aE[X] + b) = aX + b - aE[X] - b = a(X - E[X])
Therefore: Var(Y) = E[(Y - E[Y])²] = E[a²(X - E[X])²] = a²E[(X - E[X])²] = a²Var(X)
Intuition:
- Adding constant b: Shifts all values by the same amount, so relative distances (variance) remain unchanged
- Multiplying by a: Stretches/shrinks distances from the mean by factor a, but since variance uses squared distances, the effect is a²
This property shows that variance measures spread around the mean, unaffected by location shifts but sensitive to scale changes.