Methods and Systems for Implementing Market Making Strategies at Low Latency
Disclosed herein are automated trading engine embodiments that operate on market data and re-engineer trading logic to operate on computational resources that are capable of providing highly parallelized and pipelined processing operations to improve tick to trade latency. As an example, logic resources for the automated trading engine can implement multiple market making strategies that post and cancel quotes on markets. Examples of market making strategies that can be implemented by the automated trading engine include basic market making, joining market making, and bettering market making. Further still, the market making strategies can be driven by low latency derivative pricing.
1 . An apparatus for market making at low latency, the apparatus comprising:
a processor having parallelized computational resources that implement a plurality of market making strategies, wherein the marketing making strategies include at least two members of the group consisting of (1) basic market making, (2) joining market making, and (3) bettering market making;
wherein the processor is configured to receive (1) streaming underlying market data, the underlying market data pertaining to a plurality of financial instruments underlying a plurality of derivatives, (2) streaming derivatives market data, the derivatives market data comprising pricing pertaining to the derivatives, (3) a plurality of pricing parameters pertaining to the derivatives, and (4) a plurality of configuration parameters for operational control of the market making strategies; and
wherein the parallelized computational resources are configured to generate a plurality of quotes for one or more of the derivatives in accordance with the market making strategies based on a plurality of conditions applied to the streaming underlying market data, the streaming derivatives market data, the pricing parameters, and the configuration parameters, wherein the generated quotes are for transmission to one or more derivatives trading venues.
2 . The apparatus of claim 1 wherein the parallelized computational resources comprise a plurality of logic resources that operate in parallel with each other to implement the market making strategies, wherein the logic resources implement different portions of the market making strategies.
3 . The apparatus of claim 2 wherein the market making strategies include a combination of the basic and the joining market making strategies.
4 . The apparatus of claim 2 wherein the market making strategies include a combination of the basic and the bettering market making strategies.
5 . The apparatus of claim 2 wherein the market making strategies include a combination of the joining and bettering market making strategies.
6 . The apparatus of claim 2 wherein the market making strategies include a combination of the basic, joining, and bettering market making strategies.
7 . The apparatus of claim 2 wherein the market making strategies are capable of applying different pricing strategies to different sides of the generated quotes.
8 . The apparatus of claim 7 wherein the logic resources are further configured to adjust the best bid price or the best offer price for a generated quote of a subject derivative to be a defined number of tick size increments outside a subject exclusion boundary if a market making strategy would have otherwise blocked generation of the quote for having pricing that falls within the subject exclusion boundary.
9 . The apparatus of claim 1 wherein the processor is further configured to control which of the market making strategies are used to generate quotes based on a plurality of criteria.
10 . The apparatus of claim 9 wherein the criteria include derivative-specific criteria.
11 . The apparatus of claim 9 wherein the criteria include trading venue-specific criteria.
12 . The apparatus of claim 2 wherein the streaming underlying market data includes a plurality of new events for the financial instruments underlying the derivatives, wherein the logic resources are configured to compute theoretical fair prices for derivatives impacted by the new events at a rate that supports a tick to trade latency less than 1 microsecond, and wherein the generated quotes are based on the computed theoretical fair prices.
13 . The apparatus of claim 12 wherein the logic resources include a plurality of parallel logic instances that compute the theoretical fair prices for derivatives impacted by the new events in parallel.
14 . The apparatus of claim 12 wherein each parallel logic instance computes a theoretical fair price for a subject derivative impacted by a new event based on (1) first data within the streaming underlying market data that pertains to the financial instrument underlying the subject derivative, (2) second data within the streaming pricing parameters that pertains to the subject derivative, and (3) third data within the received configuration parameters that pertain to the subject derivative.
15 . The apparatus of claim 12 wherein the market making strategies include joining market making and/or bettering market making, and wherein the computed theoretical fair prices define exclusion boundaries that govern conditions for decisions to generate joining and/or bettering quotes.
16 . The apparatus of claim 15 wherein the market making strategies include bettering market making, wherein the configuration parameters include parameters that define a number of tick size increments by which the generated bettering quotes will better an existing quote, and wherein the logic resources include logic that defines conditions for the bettering market making strategy to adjust pricing for the generated quotes by the defined number of tick size increments.
17 . The apparatus of claim 12 wherein the market making strategies include basic market making, and wherein the computed theoretical fair prices define best bid and offer (BBO) pricing for a plurality of the generated quotes in combination with defined offsets for the generated quotes to establish minimum spreads for the generated quotes.
18 . The apparatus of claim 2 wherein the market making strategies include joining market making, wherein the logic resources are configured to generate a quote for a subject derivative that matches pricing for an existing market quote for the subject derivative if the pricing for the existing market quote does not fall within an exclusion boundary for the subject derivative as defined by the logic resources, and wherein the logic resources are further configured to define the exclusion boundary for the subject derivative based on a generation of theoretical fair prices for the subject derivative.
19 . The apparatus of claim 2 wherein the market making strategies include joining market making, wherein the logic resources are configured to cancel an existing quote for a quoted derivative if updated current pricing for the quoted derivative as reflected in the streaming derivatives market data falls within an exclusion boundary for the subject derivative as defined by the logic resources, and wherein the logic resources are further configured to define the exclusion boundary for the quoted derivative based on a generation of theoretical fair prices for the quoted derivative.
20 . The apparatus of claim 2 wherein the market making strategies include bettering market making, wherein the logic resources are configured to generate a quote for a subject derivative that betters pricing for an existing market quote for the subject derivative by a defined number of tick size increments if the bettered pricing for the existing market quote does not fall within an exclusion boundary for the subject derivative as defined by the logic resources, and wherein the logic resources are further configured to define the exclusion boundary for the subject derivative based on a generation of theoretical fair prices for the subject derivative.
21 . The apparatus of claim 2 wherein the market making strategies include bettering market making, wherein the logic resources are configured to cancel an existing quote for a quoted derivative if updated current pricing for the quoted derivative as reflected in the streaming derivatives market data falls within an exclusion boundary for the subject derivative as defined by the logic resources, and wherein the logic resources are further configured to define the exclusion boundary for the quoted derivative based on a generation of theoretical fair prices for the quoted derivative.
22 . The apparatus of claim 2 further comprising a quoting boundary memory that is configured to store data defining exclusion boundaries for a plurality of derivatives.
23 . The apparatus of claim 22 wherein the market making strategies include joining market making, wherein the logic resources are further configured to generate a quote for a subject derivative with pricing that matches a new market quote for the subject derivative from the streaming derivatives market data if the new market quote has pricing that falls outside the exclusion boundary in the quoting boundary memory for the subject derivative.
24 . The apparatus of claim 22 wherein the market making strategies include bettering market making, wherein the logic resources are further configured to generate a quote for a subject derivative with pricing that betters a new market quote for the subject derivative from the streaming derivatives market data by a defined number of tick size increments if the bettered new market quote has pricing that falls outside the exclusion boundary in the quoting boundary memory for the subject derivative.
25 . The apparatus of claim 1 wherein the processor comprises a field programmable gate array (FPGA), and wherein the parallelized computational resources are implemented as a hardware logic engine on the FPGA.
26 . The apparatus of claim 1 wherein the processor comprises an application-specific integrated circuit (ASIC), and wherein the parallelized computational resources are implemented as a hardware logic engine on the ASIC.
27 . The apparatus of claim 1 wherein the processor comprises a graphics processing unit (GPU), and wherein the parallelized computational resources are implemented on the GPU.
28 . The apparatus of claim 1 wherein the parallelized computational resources are further configured to analyze a plurality of the generated quotes in parallel to prioritize the quotes for scheduled transmission to a trading venue.
29 . The apparatus of claim 1 wherein the generated quotes include single-sided quotes and/or double-sided quotes.
30 . A method for market making at low latency, the method comprising:
receiving streaming underlying market data, the underlying market data pertaining to a plurality of financial instruments underlying a plurality of derivatives;
receiving streaming derivatives market data, the derivatives market data comprising pricing pertaining to the derivatives;
receiving a plurality of pricing parameters pertaining to the derivatives;
receiving a plurality of configuration parameters for operational control of a market making engine deployed in parallelized computational resources; and
the parallelized computational resources implementing a plurality of market making strategies to generate a plurality of quotes for one or more of the derivatives based on a plurality of conditions applied to the streaming underlying market data, the streaming derivatives market data, the pricing parameters, and the configuration parameters, wherein the generated quotes are for transmission to one or more derivatives trading venues, and wherein the market making strategies include at least two members of the group consisting of (1) basic market making, (2) joining market making, and (3) bettering market making.