The principle of Relative Change is fundamental to index numbers, as they focus on ratios rather than absolute differences. This allows for the comparison of variables measured in different units, such as comparing the price change of wheat (per ton) with the price change of milk (per liter) in a single composite index.
Weighting is the process of assigning importance to different items in a composite index based on their significance in the overall economy or budget. Without weights, a small change in the price of a rare luxury item would impact the index as much as a change in the price of a staple food, leading to misleading conclusions.
The Price Relative formula, , serves as the building block for more complex indices. Here, represents the price in the current period and represents the price in the base period, providing a simple percentage of change for a single item.
Simple Aggregate Method: This unweighted approach sums the prices of all items in the current year and divides them by the sum of prices in the base year. While easy to calculate, it is often criticized because it gives more weight to items with higher absolute prices, regardless of their actual consumption levels.
Laspeyres Price Index: This weighted method uses base-period quantities () as weights for both the numerator and denominator. The formula is , which is useful for comparing how the cost of a fixed 'basket' of goods from the past has changed today.
Paasche Price Index: Unlike Laspeyres, this method uses current-period quantities () as weights, calculated as . This reflects modern consumption patterns but requires new quantity data for every period, making it more difficult and expensive to maintain.
Fisher's Ideal Index: This is the geometric mean of the Laspeyres and Paasche indices, expressed as . It is considered 'ideal' because it compensates for the upward bias of Laspeyres and the downward bias of Paasche, while satisfying several mathematical tests of adequacy.
The choice between Laspeyres and Paasche indices often depends on the objective of the study and the availability of data. Laspeyres is more common in official statistics like the Consumer Price Index (CPI) because it only requires quantity data from the base year, which is easier to collect.
| Feature | Laspeyres Index | Paasche Index |
|---|---|---|
| Weights Used | Base Year Quantities () | Current Year Quantities () |
| Bias | Tends to overestimate price increases | Tends to underestimate price increases |
| Data Requirement | Easier (Base quantities fixed) | Harder (Requires new quantities each period) |
| Logic | Cost of a fixed past basket | Cost of the current basket |
Weighted Average of Price Relatives vs. Weighted Aggregative Method: The former calculates the index by taking the average of individual price relatives weighted by their value in the base year (). Both methods yield the same result if the weights are consistent, but the price relative approach is often preferred when analyzing the contribution of specific items to the total change.
Verify the Base Year: Always identify which year is the base () and which is the current () before starting calculations. A common mistake is swapping these values, which results in an index that measures the inverse of the intended change.
Check the Units: Ensure that prices and quantities are in consistent units across both periods. If one item is priced per kilogram in the base year but per gram in the current year, the data must be normalized before applying any index formula.
Sanity Check: If prices have generally risen, your index should be greater than 100. If you calculate a value like 0.85, you likely forgot to multiply by 100 or have inverted the fraction in your formula.
Fisher's Calculation: When calculating Fisher's Ideal Index, it is often faster to calculate and separately first. Then, multiply them and take the square root; this multi-step approach reduces the likelihood of complex calculator entry errors.