MODEL ACCURACY
| Collection | International Encyclopedia of Systems and Cybernetics |
|---|---|
| Year | 2004 |
| Vol. (num.) | 2(2) |
| ID | ◀ 2154 ▶ |
| Object type | Methodology or model |
H.C. GAUCH writes: “Many researchers believe that a model can be no more accurate than the data it uses. But is this so? The answer depends on three matters: the precise question being asked of the model; the design of the experiment and the quantity and accuracy of the available data”.
GAUCH gives the example of the MENDEL genetic experiences: “When he combined his data from all seven experiments, he found a ratio of 2,98:1 for the dominant and recessive traits. He modeled this as a 3:1 ratio, which also explained and predicted other experimental results” (1993, p.468). GAUCH concludes that “a model can be more accurate than the data used to build it because it amplifies hidden patterns and discards unwanted noise”.
This is a very interesting feature, quite important for systemics. It should however be taken with care: discarding “unwanted” noise could lead to fudging.