Index Calculation¶
This tab computes price indices from the balanced panel. There are two families: bilateral indices (comparing exactly two periods) and multilateral indices (using the whole panel).
Bilateral indices¶
Bilateral indices compare a base period to a comparison period. Select both periods from the dropdowns, choose a method, and click Calculate.
The result is a single index value (the price level in the comparison period relative to the base period, where base = 1.0).
Available methods¶
| Method | Weighted? | Description |
|---|---|---|
| Jevons | No | Geometric mean of price relatives. |
| Carli | No | Arithmetic mean of price relatives. |
| Dutot | No | Ratio of arithmetic mean prices. |
| Laspeyres | :material-check: | Base-period quantity basket. |
| Paasche | :material-check: | Current-period quantity basket. |
| Fisher | :material-check: | Geometric mean of Laspeyres and Paasche. |
| Törnqvist | :material-check: | Weighted geometric mean with average expenditure shares. |
| Walsh | :material-check: | Fixed basket using geometric-mean quantities. |
Weighted methods require quantity
If no quantity column was imported, the weighted methods (Laspeyres, Paasche, Fisher, Törnqvist, Walsh) are blocked. Use an unweighted method instead, or re-import data with quantities.
Exactly two periods
A bilateral calculation uses only the two selected periods. You can repeat the calculation with different period pairs.
Multilateral indices¶
Multilateral indices use the entire balanced panel to produce a full index series — one index value per period, relative to the first (chronological) period (base = 1.0).
Available methods¶
| Method | Needs quantity? | Notes |
|---|---|---|
| GEKS-Jevons | No | Unweighted; ideal for web-scraped data without quantities. |
| GEKS-Fisher | :material-check: | GEKS over the Fisher bilateral formula. |
| GEKS-Törnqvist | :material-check: | GEKS over the Törnqvist bilateral formula. |
| Geary-Khamis | :material-check: | Iterative; exposes advanced parameters (below). |
| Time Product Dummy | Optional | Supports a weighted/unweighted toggle. |
Advanced parameters¶
Geary-Khamis is an iterative method. You can tune its convergence:
| Parameter | Default | Meaning |
|---|---|---|
| Max iterations | 100 |
Upper bound on the number of iterations. |
| Tolerance | 1e-8 |
Convergence threshold. |
Time Product Dummy has a Weighted switch:
- Weighted (on) — uses expenditure shares; requires a quantity column.
- Unweighted (off) — uses prices only; works without quantities.
Result¶
The result table shows period and the computed index_value for every period.
Continue to Export to save your results.