Electronic Trading[z]
High-Touch Model
- Clients(fund managers) send order to sales trader. This can be by means of phone calling, email, FIX message or data exchanged through third party software.
- Sales Trader enters the orders into firms Order Management System(if its not electronically sent in) and assign them to different execution traders.
- Execution Trader(or Dealer) picks up the orders(and communicates to Sales Trader if necessary) and enter the order into Trader Workstation.
- The orders sent by Trader Workstation and get traded in exchange.
functional blocks. Some software system(such as Fidessa) combines both Sales Application and Dealer Trader Workstation into one GUI application.
(Click to view larger image)
The above flow is usually called High-Touch flow it requires broker firm to
provide human intervention to the client orders.
Electronic Trading(Low Touch)
Lately the concept of Electronic Trading has been popular and many broker firms have built the systems based on this concept to meet the changing client requirements:
- Clients want more direct control on the orders in the exchange
- Clients want low latency trading on the competitive market
- Clients want the service of Algorithmic trading.
The diagram below shows a typical Electronic Trading System
(Click to view larger image)
Some details about the flow:
CLIENT SENDS ORDER THROUGH FIX PROTOCOL
The Client software can be from any vendor or house built, as long as it talks FIX protocol. Broker firm should discuss the support FIX versions with Client and usually some conformance tests should be done before going live.
The broker FIX gateway receives the order and forwards it to FIX Router
FIX ROUTER
FIX Router needs to make a decision on where the order should be forward to. Subject to the system configuration and broker office setup, the FIX Gateway could be regional and OMS partitioning could be office or country specific. FIX Router needs to look at the instrument and client instruction to route order to specific OMS for next handling.
ELECTRONIC DESK HANDLING.
An electronic order can go out to market without any human intervention from broker firm.
However, just like anything in life, it can go complicate:). Orders may run into some error state or failed Criteria Checks(Criteria Check is discussed in next section) due to various reasons, such as Client sends incorrect fields including unreasonable price/quantity, or even a bug in broker’s system itself.
This is when traders in Electronic Desk need to manually sort out the error by contacting Client if the error is in the order itself or broker firm’s support staff if the problem is in the broker firm system.
ELECTRONIC TRADING OMS
In addition to the traditional OMS functionalities which requires
- Showing the client order details
- Showing executions
- Operating on orders
An Electronic Trading OMS also needs to be able to perform
- Criteria Check
- Exception reporting and handling
- Order Routing
Criteria Check
Criteria Check is to prevent trading risk and meet the compliance requirement from stock exchange regulatory.
Exception Reporting and Handling
When an order has failed the criteria check or caused some internal system error, Electronic Desk needs to be informed immediately such that they can decide to inform client or remedy the problem in a timely fashion.
The last thing a broker firm wants to happen is to let the client find out the problem before the broker firm do, which surely will upset the client and lose the firm's credibility. This is why Exception reporting and handling are important that can made or break the whole electronic trading system.
Presentation of the Exception:
- GUI alert to Electronic Desk
- Monitoring software notification to system support staff
Severity of Exception
- Info
- Warning
- Error
DMA VS. DSA
A DMA order goes straight into the market without any changes to the client order. DSA order will be routed to Algo Server for strategy execution. Subject to the algo strategies offered by the broker firm, a DSA order can result in many “slices” of child orders to the market.
Electronic Trading[z]的更多相关文章
- Algorithmic Trading[z]
Algorithmic Trading has been a hot topic for equity/derivative trading over a decade. Many ibanks an ...
- 10 Examples of HotSpot JVM Options in Java[z]
There are hundreds of JVM parameters or JVM Options exists inside sun JDK and its virtually impossib ...
- Financial Information Exchange (FIX) Protocol Interview Questions Answers[z]
What do you mean by Warrant?Warrant is a financial product which gives right to holder to Buy or Sel ...
- UNIX command Questions Answers asked in Interview
UNIX or Linux operating system has become default Server operating system and for whichever programm ...
- Some SQL basics
1, Index An index is a set of data pointers stored on disk associated with a single table. The main ...
- 理解Java多态
多态又称Polymophism,poly意思为多,polymophism即多种形态的意思.一种类型引用因为指向不同的子类,表现出不同的形态,使用不同的方法. 什么是多态 多态建议我们编码时使用comm ...
- 200个最常见的JAVA面试问题(附答案)
本文内容: 20个最常见的JAVA面试问题(附答案) 13个单例模式JAVA面试问题(附答案) 说说JVM和垃圾收集是如何工作的(附答案) 说说如何避免JAVA线程死锁(附答案) Java中HashS ...
- 将java中数组转换为ArrayList的方法实例(包括ArrayList转数组)
方法一:使用Arrays.asList()方法 1 2 String[] asset = {"equity", "stocks", "gold&q ...
- qconshanghai2017
https://2017.qconshanghai.com/schedule 第一天 (2017/10/17 星期二) 时间 日程 07:45-09:00 签到 上午 主题演讲 软件质量优化与平台创新 ...
随机推荐
- php localeconv() 函数实例讲解
php localeconv() 函数返回一包含本地数字及货币格式信息的数组.本文章向码农介绍php localeconv() 函数的使用方法和基本实例.需要的码农可以参考一下. 定义和用法 loca ...
- [Python] Scipy and Numpy(1)
import numpy as np #Create an array of 1*10^7 elements arr = np.arange(1e7) #Converting ndarray to l ...
- 讨论Android开发中的MVC设计思想
最近闲着没事,总是想想做点什么.在时间空余之时给大家说说MVC设计思想在Android开发中的运用吧! MVC设计思想在Android开发中一直都是一套比较好的设计思想.很多APP的设计都是使用这套方 ...
- python中键值叫唤例子
>>> myDict = {'a':'A','b':'B','c':'C'} >>> myDict {'a': 'A', 'c': 'C', 'b': 'B'} & ...
- leetcode976
public class Solution { public int LargestPerimeter(int[] A) { var list = A.OrderByDescending(x => ...
- net send 换行和发送广播
net send ip message 要换行的时候按ctrl+t.最后按enter 或 ctrl + m 发出 如果是批处理里面要用: 在命令行下使用:echo ^T > a.txt,注意这里 ...
- MS sql 无法进行事务日志备份
问题出在:恢复模型没有设置好我们之所以要备份数据库和事务日志都是为了以防万一,用来恢复.还原数据库的,因此恢复模型必须设置好,在sql server中默认的恢复模型为"简单".一般 ...
- 机器学习入门-概率阈值的逻辑回归对准确度和召回率的影响 lr.predict_proba(获得预测样本的概率值)
1.lr.predict_proba(under_text_x) 获得的是正负的概率值 在sklearn逻辑回归的计算过程中,使用的是大于0.5的是正值,小于0.5的是负值,我们使用使用不同的概率结 ...
- ABAP-HTTP发送JSON
1. HTTP发送JSON格式数据: function zap_01_url_sap_to_c3 . *"------------------------------------------ ...
- Oracle12c部署
部署环境业务系统与数据库服务部署在一台服务器上了 电脑是台式机没有网络,也没有插网线,需要先建立一个网络回环,然后进行Oracle12c的安装,安装过程中系统会默认勾选创建为容器数据库,需要把这个勾选 ...