Understanding Types of Means | Set 1

Table of Contents#

1. Arithmetic Mean (AM)

   • Definition & Formula
   • Example Usage
   • When to Use AM
   • Common Practices & Pitfalls
   • Best Practices

2. Geometric Mean (GM)

   • Definition & Formula
   • Example Usage
   • When to Use GM
   • Common Practices & Pitfalls
   • Best Practices

3. Harmonic Mean (HM)

   • Definition & Formula
   • Example Usage
   • When to Use HM
   • Common Practices & Pitfalls
   • Best Practices

4. Weighted Arithmetic Mean (WAM)

   • Definition & Formula
   • Example Usage
   • When to Use WAM
   • Common Practices & Pitfalls
   • Best Practices

5. Summary & Key Takeaways

6. References

1. Arithmetic Mean (AM)#

Definition & Formula#

The Arithmetic Mean (AM) is the most familiar type of mean. It is calculated by summing all values in a dataset and dividing by the number of values.

Formula:
For a dataset ( x_1, x_2, ..., x_n ), the arithmetic mean is:
[ \bar{x} = \frac{x_1 + x_2 + ... + x_n}{n} = \frac{\sum_{i=1}^{n} x_i}{n} ]
where ( n ) is the number of observations, and ( \sum_{i=1}^{n} x_i ) denotes the sum of all values.

Example Usage#

Suppose you want to find the average test score of 5 students: 85, 90, 78, 92, and 88.

Calculation:
[ \bar{x} = \frac{85 + 90 + 78 + 92 + 88}{5} = \frac{433}{5} = 86.6 ]
The average test score is 86.6.

When to Use AM#

• Symmetric Data: AM works best for datasets with a symmetric distribution (e.g., heights, test scores), where values cluster around a central point.
• Additive Relationships: Use AM when the data represents additive quantities (e.g., total rainfall over days, average monthly income).

Common Practices & Pitfalls#

• Everyday Averages: AM is the "default" mean for most casual use cases (e.g., average temperature, average speed over time).
• Sensitivity to Outliers: AM is highly influenced by extreme values (outliers). For example, if one student scored 150 (an outlier) in the test example above, the AM becomes ( \frac{433 + 150}{6} = 97.17 ), which no longer represents the "typical" score.

Best Practices#

• Check for Outliers: Always visualize data (e.g., box plots, histograms) to identify outliers before using AM. If outliers exist, consider the median instead.
• Normalize Units: Ensure all values are in the same units (e.g., convert "hours" and "minutes" to a single unit) to avoid skewed results.

2. Geometric Mean (GM)#

Definition & Formula#

The Geometric Mean (GM) is used for datasets involving multiplicative growth, ratios, or rates (e.g., investment returns, population growth). It is calculated as the ( n )-th root of the product of all values.

Formula:
For a dataset ( x_1, x_2, ..., x_n ) (where all ( x_i > 0 )), the geometric mean is:
[ GM = \sqrt[n]{x_1 \times x_2 \times ... \times x_n} = \left( \prod_{i=1}^{n} x_i \right)^{1/n} ]
where ( \prod_{i=1}^{n} x_i ) denotes the product of all values.

Example Usage#

Suppose you invest $1000 with annual returns of 5%, 10%, -2%, and 8% over 4 years. What is the average annual growth rate?

Step 1: Convert percentages to growth multipliers:
5% → 1.05, 10% → 1.10, -2% → 0.98, 8% → 1.08.

Step 2: Compute the product of multipliers:
( 1.05 \times 1.10 \times 0.98 \times 1.08 \approx 1.235 ).

Step 3: Take the 4th root (since ( n=4 )):
( GM = \sqrt[4]{1.235} \approx 1.054 ).

Step 4: Convert back to percentage:
Average annual growth rate = ( (1.054 - 1) \times 100 = 5.4% ).

When to Use GM#

• Multiplicative Growth: Use GM for data involving compounding (e.g., investment returns, population growth rates, bacterial growth).
• Ratios and Proportions: GM is ideal for averaging ratios (e.g., price/earnings ratios, exchange rates).

Common Practices & Pitfalls#

• Positive Values Only: GM is undefined for zero or negative values (since the root of a negative number is not real).
• Logarithmic Simplification: For large datasets, compute GM using logarithms to avoid numerical overflow:
  [ GM = e^{\frac{1}{n} \sum_{i=1}^{n} \ln(x_i)} ]

Best Practices#

• Normalize to Multipliers: For growth rates, always convert percentages to multipliers (e.g., 5% → 1.05) before calculating GM.
• Avoid Zero/Negative Values: If data contains zeros or negatives, GM is invalid—use AM or another mean instead.

3. Harmonic Mean (HM)#

Definition & Formula#

The Harmonic Mean (HM) is used to average rates, speeds, or ratios where the denominator of the ratio is constant (e.g., average speed for a round trip). It is the reciprocal of the arithmetic mean of reciprocals.

Formula:
For a dataset ( x_1, x_2, ..., x_n ) (where all ( x_i > 0 )), the harmonic mean is:
[ HM = \frac{n}{\frac{1}{x_1} + \frac{1}{x_2} + ... + \frac{1}{x_n}} = \frac{n}{\sum_{i=1}^{n} \frac{1}{x_i}} ]

Example Usage#

You drive 100 km at 60 km/h and another 100 km at 40 km/h. What is your average speed for the entire trip?

Step 1: Identify distances and speeds:
Distance 1 = 100 km, Speed 1 = 60 km/h; Distance 2 = 100 km, Speed 2 = 40 km/h.

Step 2: Time for each segment:
Time 1 = ( \frac{100}{60} = \frac{5}{3} ) hours; Time 2 = ( \frac{100}{40} = \frac{5}{2} ) hours.

Step 3: Total distance = 200 km; Total time = ( \frac{5}{3} + \frac{5}{2} = \frac{25}{6} ) hours.

Step 4: Average speed = ( \frac{\text{Total Distance}}{\text{Total Time}} = \frac{200}{\frac{25}{6}} = 48 ) km/h.

Using HM:
[ HM = \frac{2}{\frac{1}{60} + \frac{1}{40}} = \frac{2}{\frac{2 + 3}{120}} = \frac{2 \times 120}{5} = 48 \text{ km/h}. ]

When to Use HM#

• Rates with Constant Denominator: Use HM for averaging rates where the "denominator" (e.g., distance, time) is fixed (e.g., average speed for equal distances, average fuel efficiency).
• Inverse Relationships: HM is useful when data exhibits an inverse relationship (e.g., price per unit, where lower prices imply higher quantity).

Common Practices & Pitfalls#

• Equal Weights: HM assumes equal "weight" for each rate (e.g., equal distance in the speed example). For unequal weights, use the weighted harmonic mean.
• Sensitivity to Small Values: HM is pulled toward smaller values more than AM. For example, HM of 1, 2, 3 is ( \frac{3}{1 + 0.5 + 0.333} \approx 1.636 ), which is lower than AM (2).

Best Practices#

• Check Units: Ensure all rates are in the same units (e.g., km/h, not a mix of km/h and m/s).
• Avoid Zeros: HM is undefined if any ( x_i = 0 ) (division by zero).

4. Weighted Arithmetic Mean (WAM)#

Definition & Formula#

The Weighted Arithmetic Mean (WAM) accounts for the "importance" or "weight" of each data point. Unlike the regular arithmetic mean (which assigns equal weight to all values), WAM assigns higher weights to more significant observations.

Formula:
For a dataset ( x_1, x_2, ..., x_n ) with weights ( w_1, w_2, ..., w_n ), the weighted arithmetic mean is:
[ WAM = \frac{w_1 x_1 + w_2 x_2 + ... + w_n x_n}{w_1 + w_2 + ... + w_n} = \frac{\sum_{i=1}^{n} w_i x_i}{\sum_{i=1}^{n} w_i} ]

Example Usage#

A student’s course grade is determined by exams (40% weight), quizzes (30% weight), and homework (30% weight). If they scored 85 on exams, 90 on quizzes, and 75 on homework, what is their final grade?

Calculation:
[ WAM = \frac{(0.4 \times 85) + (0.3 \times 90) + (0.3 \times 75)}{0.4 + 0.3 + 0.3} = \frac{34 + 27 + 22.5}{1} = 83.5 ]

When to Use WAM#

• Unequal Importance: Use WAM when some data points are more critical than others (e.g., course grades with weighted assignments, portfolio returns with asset weights).
• Aggregated Data: WAM is useful for combining averages from subgroups with different sizes (e.g., average salary across departments with varying employee counts).

Common Practices & Pitfalls#

• Normalizing Weights: Weights often sum to 1 (e.g., percentages) for interpretability, but they can also be raw counts (e.g., number of employees in a department).
• Subjective Weights: Weights must be defined based on objective criteria (e.g., credit hours for courses) to avoid bias.

Best Practices#

• Document Weights: Clearly define how weights are chosen (e.g., "Exam weight = 40% based on syllabus guidelines").
• Normalize if Needed: If weights do not sum to 1, divide by the total weight to ensure the mean is interpretable.

5. Summary & Key Takeaways#

6. References#

• DeGroot, M. H., & Schervish, M. J. (2012). Probability and Statistics (4th ed.). Pearson.
• Khan Academy. (n.d.). Means: Arithmetic, Geometric, and Harmonic. https://www.khanacademy.org
• Wikipedia. (2023). Mean. https://en.wikipedia.org/wiki/Mean
• Statistics How To. (n.d.). Harmonic Mean: Definition, Formula, and Examples. https://www.statisticshowto.com