Do Tick Volumes Work in Forex?

本文探讨了TickVolume在外汇市场中的应用及其实用性。通过对比TickVolume与货币成交量之间的相关性,研究发现两者间存在较强的相关系数。实验中使用了来自不同交易平台的数据进行比较,验证了TickVolume作为有效分析工具的可靠性。

摘要生成于 C知道 ,由 DeepSeek-R1 满血版支持, 前往体验 >

Do Tick Volumes Work in Forex?

Tick volumes in Forex have sparked much debate among trading communities on what information they provide, and how they compare to regular volumes. In this article we will see whether tick volumes can be reliably used in Forex.

According to high-profile forex data published in “The Microstructure Approach to Exchange Rates” by Richard Lyons (page 201), the correlation coefficient between tick and money volume equals 0.98. Tick volumes being the number of trades, and money volumes being the net amount of bought or sold units. However this coefficient dates back to 1998, and we are interested in finding the correlation coefficient (CC) at a more recent time.

 

Let’s take hourly data for 2013 and include three types of volumes:

  • Tick volume – we export it from Metatrader 4 connected to Alpari
  • ECN – exported from Dukascopy website
  • Futures volumes – exported from EUR and USD futures

I limit the experiment only to 2013 because of a limitation of available data for futures. However, you can download the data for almost any period from Alpari and Dukascopy.

 

Downloading the data


Dukascopy ECN Volumes

Dukascopy ECN Volumes

 

Alpari Tick Volumes

Alpari Tick Volumes

 

Futures Volumes

Futures Volumes

We don’t need the OHLC. Now let’s combine all the necessary data into one spreadsheet.

 

The final spreadsheet has 5284 bars. Those are the bars that coincide in all the three files.

Volumes comparison

Volumes comparison

 

And now we can calculate the correlation coefficient:

Correlation Coefficient

Correlation Coefficient

 

Results

After all the calculations we can see the following coefficients:

  • CC between the futures volumes and the tick volumes (Alpari) equals 0.91
  • CC between the futures volumes and the ECN volumes (Dukascopy) equals 0.79
  • CC between the ECN volumes (Dukascopy) and the tick volumes (Alpari) equals 0.85

All the three numbers are in excess of 0.75. This confirms the strong correlation between all the three types of volumes, and we can freely substitute one measure for the other in our analysis.

 

In conclusion I’d like to note that all the coefficients are less than 0.98, in no way means that the interdependence between the volumes is trending down. The reason is elsewhere: the 1998’s data compared EBS (electronic broker system) tick volumes directly to money volume, whereas we used tick volumes from individual brokers (Dukascopy and Alpari).

 

Author: Roman C.

Date: 23/August/2016

内容概要:本文详细探讨了基于樽海鞘算法(SSA)优化的极限学习机(ELM)在回归预测任务中的应用,并与传统的BP神经网络、广义回归神经网络(GRNN)以及未优化的ELM进行了性能对比。首先介绍了ELM的基本原理,即通过随机生成输入层与隐藏层之间的连接权重及阈值,仅需计算输出权重即可快速完成训练。接着阐述了SSA的工作机制,利用樽海鞘群体觅食行为优化ELM的输入权重和隐藏层阈值,从而提高模型性能。随后分别给出了BP、GRNN、ELM和SSA-ELM的具体实现代码,并通过波士顿房价数据集和其他工业数据集验证了各模型的表现。结果显示,SSA-ELM在预测精度方面显著优于其他三种方法,尽管其训练时间较长,但在实际应用中仍具有明显优势。 适合人群:对机器学习尤其是回归预测感兴趣的科研人员和技术开发者,特别是那些希望深入了解ELM及其优化方法的人。 使用场景及目标:适用于需要高效、高精度回归预测的应用场景,如金融建模、工业数据分析等。主要目标是提供一种更为有效的回归预测解决方案,尤其是在处理大规模数据集时能够保持较高的预测精度。 其他说明:文中提供了详细的代码示例和性能对比图表,帮助读者更好地理解和复现实验结果。同时提醒使用者注意SSA参数的选择对模型性能的影响,建议进行参数敏感性分析以获得最佳效果。
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值