The effect of algorithmic trading on agricultural commodities market quality
Bibliographic record
Abstract
The growing use of algorithms has significantly changed trading. These changes have been subjected to an ongoing debate in the finance literature. Some studies have found that algorithmic trading (AT) has a positive effect on market quality by increasing the competition, trading volume and liquidity, and lowering trading costs. Algorithmic traders provide liquidity when it is expensive and take it when it is cheap. On the other hand, others argue that AT may increase volatility and adverse selection. The difference in speed between fast and slow traders not only causes adverse selection, but it leads to wider spreads. We study the effect of AT on market quality, trade size and volatility, focusing on five agricultural commodity futures markets listed in the CME Group during the period of December 2015 to March 2016. The commodities include wheat, soybean, corn, lean hogs and live cattle. We control for USDA announcements released during the period of study, the day of the week, and intraday movements of AT. We find that AT improves market quality by narrowing the effective half spread (an estimate of the liquidity cost) in all markets. The effect is stronger in lean hogs and live cattle markets where AT also decreases the adverse selection (the reflection of the existence of different levels of information in the market). Algorithmic traders are more active when transaction costs and information asymmetry are lower. AT also decreases volatility in all markets. Our results show that the USDA announcements are significant only in the soybean market. We also find that the effect of the day of the week on AT is only significant in the corn market. The effect of the opening time of the market on AT is positive in soybean and corn, and negative in live cattle. The closing time is negative in all markets except live cattle where it is not significant. Finally, we perform an impulse response analysis. We find that the initial reaction of QHS and RS to a shock of AT is positive, the reaction of EHS and PI is negative, and the effect is always temporary.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.037 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".