Working Paper: CEPR ID: DP16019
Authors: Jerome Dugast; Thierry Foucault
Abstract: We analyze how computing power and data abundance affect speculators' search for predictors. Speculators search for predictors through trials and optimally stop searching when they find a predictor with a signal-to-noise ratio larger than an endogenous threshold. Greater computing power raises this threshold by reducing search costs. In contrast, data abundance can reduce this threshold because (i) it intensifies competition among speculators and (ii) it increases the average number of trials to find a predictor. We derive implications of these findings for the distribution of asset managers' skills and trading profits and the informativeness of asset prices.
Keywords: alternative data; data abundance; data mining; price informativeness; search for information
JEL Codes: No JEL codes provided
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Increased computing power (C89) | Reduced search costs (D83) |
| Reduced search costs (D83) | Higher threshold for predictors that speculators will accept (D84) |
| Higher threshold for predictors that speculators will accept (D84) | Higher quality predictors used in trading (C58) |
| Increased computing power (C89) | More stringent stopping rules for speculators (G18) |
| More stringent stopping rules for speculators (G18) | Increased informativeness of asset prices (G19) |
| Data abundance (C55) | Push back the data frontier (C55) |
| Data abundance (C55) | Exacerbate the needle in the haystack problem (C55) |
| Push back the data frontier (C55) | Increased likelihood of finding more informative predictors (C52) |
| Exacerbate the needle in the haystack problem (C55) | Lower average quality of predictors used in trading (C29) |
| Data abundance (C55) | Ambiguous effects on predictor quality (C52) |
| Greater computing power (C89) | Improved price informativeness (G19) |