Descriptive Statistics: Measures of Central Tendency and Dispersion, Correlation, Index Numbers

Welcome, future economists! Today, we dive into the foundational concepts of descriptive statistics, which are crucial for understanding and interpreting economic data. Descriptive statistics help us summarize and present data in a meaningful way. We will cover measures of central tendency, measures of dispersion, correlation, and index numbers. These tools are not just theoretical; they are used daily in economic analysis, forecasting, and policy-making.

1. Measures of Central Tendency

Measures of central tendency are statistical measures used to identify the center or typical value of a dataset. They provide a single value that represents the dataset's core or average. The most common measures are the mean, median, and mode.

1.1. Mean (Arithmetic Average)

The mean is the sum of all values in a dataset divided by the number of values. It's the most commonly used measure of central tendency.

Formula: $$ \bar{x} = \frac{\sum_{i=1}^{n} x_i}{n} $$ Where:

• $ \bar{x} $ is the sample mean
• $ x_i $ represents each individual value in the dataset
• $ n $ is the total number of values in the dataset

Example: Consider the monthly incomes of five employees: ₹20,000, ₹25,000, ₹30,000, ₹35,000, ₹40,000. The sum of incomes = 20,000 + 25,000 + 30,000 + 35,000 + 40,000 = ₹150,000. The number of employees ($n$) = 5. Mean income = ₹150,000 / 5 = ₹30,000.

Pros: Uses all data points, unique, mathematically tractable. Cons: Highly sensitive to outliers (extreme values). For instance, if one employee earned ₹1,00,000, the mean would be significantly skewed upwards.

1.2. Median

The median is the middle value in a dataset that has been ordered from least to greatest. It divides the dataset into two equal halves.

Calculation:

• If the number of observations ($n$) is odd, the median is the middle value. The position is $ (n+1)/2 $.
• If the number of observations ($n$) is even, the median is the average of the two middle values. The positions are $ n/2 $ and $ (n/2) + 1 $.

Example (Odd $n$): Using the incomes ₹20,000, ₹25,000, ₹30,000, ₹35,000, ₹40,000. The dataset is already ordered. $n=5$ (odd). The median position is $ (5+1)/2 = 3 $. The 3rd value is ₹30,000. Median income = ₹30,000.

Example (Even $n$): Consider incomes: ₹20,000, ₹25,000, ₹30,000, ₹35,000, ₹40,000, ₹50,000. The dataset is ordered. $n=6$ (even). The median positions are $ 6/2 = 3 $ and $ (6/2) + 1 = 4 $. The 3rd value is ₹30,000, and the 4th value is ₹35,000. Median income = (₹30,000 + ₹35,000) / 2 = ₹32,500.

Pros: Not affected by outliers, good for skewed distributions. Cons: Does not use all data points, less mathematically convenient than the mean.

1.3. Mode

The mode is the value that appears most frequently in a dataset. A dataset can have one mode (unimodal), two modes (bimodal), or more (multimodal).

Example: Consider the number of cars sold by a dealership over 10 days: 2, 3, 5, 3, 4, 3, 2, 6, 3, 5. The values are: 2, 2, 3, 3, 3, 3, 4, 5, 5, 6. The value '3' appears 4 times, which is more than any other value. Mode = 3 cars.

Example (Bimodal): Dataset: 10, 12, 12, 13, 14, 14, 15. The values 12 and 14 both appear twice. Modes = 12 and 14.

Pros: Easy to find, applicable to non-numeric data (e.g., most frequent product color). Cons: May not exist, may not be unique, may not be close to the center of the data.

2. Measures of Dispersion (Measures of Variability)

Measures of dispersion indicate how spread out or scattered the data points are in a dataset. They tell us about the variability of the data. Key measures include range, variance, and standard deviation.

2.1. Range

The range is the difference between the highest and lowest values in a dataset.

Formula: $$ \text{Range} = \text{Maximum Value} - \text{Minimum Value} $$

Example: Using the incomes ₹20,000, ₹25,000, ₹30,000, ₹35,000, ₹40,000. Maximum Value = ₹40,000. Minimum Value = ₹20,000. Range = ₹40,000 - ₹20,000 = ₹20,000.

Pros: Simple to calculate. Cons: Only uses two extreme values, highly sensitive to outliers.

2.2. Variance

Variance measures the average squared difference of each data point from the mean. It quantifies the spread of the data.

Formula (Sample Variance, $s^2$): $$ s^2 = \frac{\sum_{i=1}^{n} (x_i - \bar{x})^2}{n-1} $$ Formula (Population Variance, $ \sigma^2 $): $$ \sigma^2 = \frac{\sum_{i=1}^{N} (x_i - \mu)^2}{N} $$ Where:

• $ \bar{x} $ or $ \mu $ is the mean of the sample or population
• $ x_i $ is each individual value
• $ n $ or $ N $ is the number of observations

We use $n-1$ in the denominator for sample variance to provide an unbiased estimate of the population variance.

Example: Using incomes ₹20,000, ₹25,000, ₹30,000, ₹35,000, ₹40,000. Mean ($ \bar{x} $) = ₹30,000. Calculate deviations from the mean: (20000 - 30000) = -10000 (25000 - 30000) = -5000 (30000 - 30000) = 0 (35000 - 30000) = 5000 (40000 - 30000) = 10000 Square the deviations: (-10000)^2 = 100,000,000 (-5000)^2 = 25,000,000 (0)^2 = 0 (5000)^2 = 25,000,000 (10000)^2 = 100,000,000 Sum of squared deviations = 100M + 25M + 0 + 25M + 100M = 250,000,000. $n = 5$. So, $n-1 = 4$. Sample Variance ($s^2$) = 250,000,000 / 4 = 62,500,000.

Pros: Uses all data points, mathematically useful. Cons: Units are squared (e.g., ₹²), making interpretation difficult. Sensitive to outliers.

2.3. Standard Deviation

The standard deviation is the square root of the variance. It is the most widely used measure of dispersion because it is in the same units as the original data, making it easier to interpret.

Formula (Sample Standard Deviation, $s$): $$ s = \sqrt{s^2} = \sqrt{\frac{\sum_{i=1}^{n} (x_i - \bar{x})^2}{n-1}} $$ Formula (Population Standard Deviation, $ \sigma $): $$ \sigma = \sqrt{\sigma^2} = \sqrt{\frac{\sum_{i=1}^{N} (x_i - \mu)^2}{N}} $$

Example: Using the variance calculated above ($s^2$ = 62,500,000). Sample Standard Deviation ($s$) = $ \sqrt{62,500,000} $ = ₹7,905.69 (approximately).

Interpretation: On average, the monthly incomes deviate from the mean income of ₹30,000 by about ₹7,905.69. A smaller standard deviation indicates data points are clustered around the mean, while a larger one indicates they are spread out.

3. Correlation

Correlation measures the strength and direction of a linear relationship between two quantitative variables. It does not imply causation. The most common measure is the Pearson correlation coefficient ($r$).

3.1. Pearson Correlation Coefficient ($r$)

The Pearson correlation coefficient ($r$) ranges from -1 to +1.

• $ r = +1 $: Perfect positive linear correlation. As one variable increases, the other increases proportionally.
• $ r = -1 $: Perfect negative linear correlation. As one variable increases, the other decreases proportionally.
• $ r = 0 $: No linear correlation.
• Values between 0 and +1 indicate positive correlation (strength increases as $r$ approaches 1).
• Values between 0 and -1 indicate negative correlation (strength increases as $r$ approaches -1).

Formula: $$ r = \frac{\sum (x_i - \bar{x})(y_i - \bar{y})}{\sqrt{\sum (x_i - \bar{x})^2 \sum (y_i - \bar{y})^2}} $$ Alternatively, using covariance and standard deviations: $$ r = \frac{\text{Cov}(x, y)}{s_x s_y} $$ Where:

• $ x_i, y_i $ are the individual data points
• $ \bar{x}, \bar{y} $ are the means of the $x$ and $y$ variables
• $ s_x, s_y $ are the standard deviations of the $x$ and $y$ variables
• $ \text{Cov}(x, y) $ is the covariance between $x$ and $y$

Example: Suppose we examine the relationship between hours studied ($x$) and exam scores ($y$) for 5 students. Data: (Hours Studied, Score): (2, 60), (3, 70), (5, 80), (6, 85), (8, 95) Calculate means: $ \bar{x} = (2+3+5+6+8)/5 = 4.4 $ hours. $ \bar{y} = (60+70+80+85+95)/5 = 78 $. Calculate deviations and products: | x | y | x - x̄ | y - ȳ | (x - x̄)(y - ȳ) | (x - x̄)² | (y - ȳ)² | |---|---|-------|-------|-----------------|----------|----------| | 2 | 60 | -2.4 | -18 | 43.2 | 5.76 | 324 | | 3 | 70 | -1.4 | -8 | 11.2 | 1.96 | 64 | | 5 | 80 | 0.6 | 2 | 1.2 | 0.36 | 4 | | 6 | 85 | 1.6 | 7 | 11.2 | 2.56 | 49 | | 8 | 95 | 3.6 | 17 | 61.2 | 12.96 | 289 | | Sum | | | | 128.0 | 23.60 | 730 | $ r = \frac{128}{\sqrt{23.60 \times 730}} = \frac{128}{\sqrt{17228}} \approx \frac{128}{131.25} \approx 0.975 $ This indicates a very strong positive linear relationship between hours studied and exam scores.

Important Note: Correlation does not imply causation. For instance, ice cream sales and crime rates might be positively correlated because both increase in hot weather, not because one causes the other.

3.2. Coefficient of Determination ($r^2$)

The coefficient of determination ($r^2$) is the square of the Pearson correlation coefficient. It represents the proportion of the variance in one variable that is predictable from the other variable.

Example: In the previous example, $r = 0.975$. $ r^2 = (0.975)^2 \approx 0.951 $. This means that approximately 95.1% of the variation in exam scores can be explained by the variation in hours studied (in a linear fashion). The remaining 4.9% is due to other factors.

4. Index Numbers

Index numbers are statistical measures that show changes in a variable or a group of related variables over time, relative to a base period. They are widely used in economics to track inflation, economic growth, and changes in consumer prices.

4.1. Simple Index Numbers

A simple index number compares the value of a single item in different periods to its value in a base period.

Formula: $$ \text{Simple Index} = \left( \frac{\text{Current Period Value}}{\text{Base Period Value}} \right) \times 100 $$

Example: The price of a loaf of bread was ₹20 in 2020 (base year) and ₹25 in 2023. Index of bread price in 2023 = (₹25 / ₹20) * 100 = 125. This means the price of bread in 2023 is 125% of its price in 2020, an increase of 25%.

4.2. Aggregate Index Numbers

Aggregate index numbers are used to measure changes in a group of related items. They can be calculated in two main ways:

• Simple Aggregate Method: Sums the prices (or quantities) of all items in the current period and divides by the sum in the base period.
• Weighted Aggregate Method: Assigns weights to different items based on their importance.

4.3. Weighted Aggregate Index Numbers

These are more common and accurate as they account for the relative importance of different items. Two popular methods are:

4.3.1. Laspeyres Index

This index uses the quantities from the *base period* as weights. It measures the cost of purchasing a fixed basket of goods from the base period in the current period.

Formula: $$ L = \frac{\sum (P_t \times Q_0)}{\sum (P_0 \times Q_0)} \times 100 $$ Where:

• $ P_t $ = Price in the current period
• $ P_0 $ = Price in the base period
• $ Q_0 $ = Quantity in the base period

Pros: Relatively easy to calculate as base period quantities are usually known. Cons: Tends to overstate price increases because it doesn't account for consumers substituting cheaper goods when prices rise (i.e., it uses a fixed, potentially outdated, basket).

4.3.2. Paasche Index

This index uses the quantities from the *current period* as weights. It measures the cost of purchasing the current period's basket of goods using current prices versus base period prices.

Formula: $$ P = \frac{\sum (P_t \times Q_t)}{\sum (P_0 \times Q_t)} \times 100 $$ Where:

• $ P_t $ = Price in the current period
• $ P_0 $ = Price in the base period
• $ Q_t $ = Quantity in the current period

Pros: More accurately reflects current consumption patterns. Cons: More difficult to calculate as current period quantities need to be collected. Tends to understate price increases.

4.3.3. Fisher's Ideal Index

This index is considered "ideal" because it is the geometric mean of the Laspeyres and Paasche indices. It overcomes some of the biases of both.

Formula: $$ F = \sqrt{L \times P} $$ It also satisfies certain desirable properties like time reversal and factor reversal tests.

4.4. Consumer Price Index (CPI)

The CPI is a widely used measure of inflation. It tracks the average change over time in the prices paid by urban consumers for a market basket of consumer goods and services. The basket is typically fixed for a period using base-year quantities (making it similar to a Laspeyres index).

Example Calculation (Simplified): Assume a basket contains only Bread and Milk. Base Year (2020): Bread price = ₹20, Quantity = 10 units. Milk price = ₹30, Quantity = 5 units. Current Year (2023): Bread price = ₹25, Quantity = 10 units. Milk price = ₹35, Quantity = 5 units.

Item	$ P_0 $	$ Q_0 $	$ P_t $	$ P_0 \times Q_0 $	$ P_t \times Q_0 $
Bread	20	10	25	200	250
Milk	30	5	35	150	175
Total				350	425

Laspeyres Index ($L$) = (425 / 350) * 100 = 121.43. This suggests that the cost of this specific basket of goods has increased by 21.43% from 2020 to 2023.