Notes on Key Statistical Topics
Log-normal, Error, DOF, and ANOVA
Your Name
Your Institution
October 10, 2025
Your Name (Your Institution) Notes on Key Statistical Topics October 10, 2025 1 / 10
Outline
1 Log-normal Statistics
2 Error and Uncertainty
3 Degrees of Freedom (DOF)
4 Analysis of Variance (ANOVA)
Your Name (Your Institution) Notes on Key Statistical Topics October 10, 2025 2 / 10
Log-normal Statistics: Notes
A random variable X is log-normally distributed if its natural
logarithm, Y = ln(X ), is normally distributed.
Useful for modeling variables that are always positive and
right-skewed (e.g., income, rainfall).
Key Characteristics
Always Positive: Values are always X > 0.
Right-Skewed: Has a long tail to the right.
Parameters: Defined by µ and σ (the mean and standard deviation
of ln(X )).
Important Formulas
σ2
Mean of X: E [X ] = e µ+ 2
Median of X: Median(X ) = e µ
2
σNotes 2
Your Name (Your Institution) on Key 2µ+σ
Statistical Topics October 10, 2025 3 / 10
Log-normal Statistics: Problems
1 If ln(X ) is normal with µ = 2 and σ = 1, what is the median of X ?
Solution: Median = e µ = e 2 ≈ 7.39
2 For the same data, what is the mean of X ?
2 /2 2 /2
Solution: Mean = e µ+σ = e 2+1 = e 2.5 ≈ 12.18
3 Pollutant concentration follows a log-normal distribution with µ = 3.5
and σ = 0.5. Find the mean concentration.
2 /2
Solution: Mean = e 3.5+(0.5) = e 3.625 ≈ 37.52
4 If the median of a log-normal variable is 20, what is µ?
Solution: 20 = e µ =⇒ µ = ln(20) ≈ 3.00
Your Name (Your Institution) Notes on Key Statistical Topics October 10, 2025 4 / 10
Error and Uncertainty: Notes
Types of Errors Definitions
Systematic Errors: Consistent, Accuracy: How close a
repeatable errors affecting measurement is to the true
accuracy. value.
Random Errors: Unpredictable Precision: How close multiple
fluctuations affecting precision. measurements are to each other.
Propagation of Uncertainty
Addition/Subtraction (Z = A ± B):
q
δZ = (δA)2 + (δB)2
Multiplication/Division (Z = A × B or A/B):
s 2 2
δZ δA δB
Your Name (Your Institution) = +Topics
Notes on Key Statistical October 10, 2025 5 / 10
Error and Uncertainty: Problems
1 You measure L1 = (10.2 ± 0.1) cm and L2 = (5.3 ± 0.2) cm. Find the
total length and its uncertainty.
√ √
Ltotal = 15.5 cm. δLtotal = 0.12 + 0.22 = 0.05 ≈ 0.22 cm. Result:
(15.5 ± 0.22) cm.
2 A resistor has V = (12.0 ± 0.3) V and I = (2.0 ± 0.1) A. Calculate
resistance R = V /I with its uncertainty.
q
R = 6.0 Ω. δR
R = ( 0.3 2 0.1 2
12 ) + ( 2 ) ≈ 0.0559.
δR = 6.0 × 0.0559 ≈ 0.34 Ω. Result: (6.0 ± 0.34) Ω.
Your Name (Your Institution) Notes on Key Statistical Topics October 10, 2025 6 / 10
Degrees of Freedom (DOF): Notes
Degrees of freedom (DOF or ν) is the number of independent
values that are free to vary in a calculation.
Analogy: If 3 numbers must sum to 10, you can freely pick the first
two, but the third is fixed. You have 3 − 1 = 2 DOF.
In statistics, we lose a degree of freedom for each parameter we
estimate from the data.
Common Scenarios
Sample Standard Deviation: df = n − 1
One-sample t-test: df = n − 1
Chi-Square Goodness-of-Fit Test: df = (number of categories − 1)
Your Name (Your Institution) Notes on Key Statistical Topics October 10, 2025 7 / 10
Degrees of Freedom: Problems
1 For a sample of 15 test scores, what is the DOF for calculating the
sample variance?
Solution: df = n − 1 = 15 − 1 = 14.
2 A t-test is performed on 25 measurements. What is the DOF?
Solution: df = n − 1 = 25 − 1 = 24.
3 A chi-square test checks if a 6-sided die is fair. What is the DOF?
Solution: df = (categories − 1) = 6 − 1 = 5.
4 An experiment has two groups: Group A (n1 = 10) and Group B
(n2 = 12). Find the total DOF for an independent t-test.
Solution: df = (n1 − 1) + (n2 − 1) = (10 − 1) + (12 − 1) = 9 + 11 = 20.
Your Name (Your Institution) Notes on Key Statistical Topics October 10, 2025 8 / 10
Analysis of Variance (ANOVA):
ANOVA is a test to find significant differences between the means
of three or more groups.
It works by comparing the variation between groups to the variation
within groups.
The Core Idea: Partitioning Variance
Between-Group Variance (MSB): How much the group means vary
from the overall mean.
Within-Group Variance (MSW ): How much data varies within
each group (the ”noise”).
The F-statistic
The ratio of the two variances. A large F-statistic suggests the group
means are different.
Variance between groups MSB
F = =
Variance within groups MSW
Your Name (Your Institution) Notes on Key Statistical Topics October 10, 2025 9 / 10
ANOVA: Problems
Given: k = # of groups, N = total observations, SSB = Sum of Squares
Between, SSW = Sum of Squares Within.
1 For k = 3, N = 15, SSB = 80, SSW = 120, calculate MSB.
dfbetween = k − 1 = 2. MSB = SSB/dfbetween = 80/2 = 40.
2 Using the same data, calculate MSW .
dfwithin = N − k = 12. MSW = SSW /dfwithin = 120/12 = 10.
3 From the results above, find the F-statistic.
F = MSB/MSW = 40/10 = 4.0.
4 A study with k = 4, N = 40 finds MSB = 90 and MSW = 15.
Calculate the F-statistic.
F = MSB/MSW = 90/15 = 6.0.
Your Name (Your Institution) Notes on Key Statistical Topics October 10, 2025 10 / 10