Forex Trading – Perspective from an Index Future Trader (Part 2)
There are some more interesting differences between Forex trading and stock/index future trading. I will continue my discussion in this second instalment.
24 Hours Trading
Forex trading starts every week from Sunday to Friday. The starting time of trading on Sunday varies among the forex brokerages, so is the ending time on Friday.
Some brokerages allow you to place orders on Sunday as early as 1 PM Eastern Time in the afternoon while some others start at a later time like 5 PM to 6 PM Eastern Time. For the ending time on Friday though, it is more consistent at 4 PM to 6 PM Eastern Time.
In between the starting time and the ending time, it is 24 hours trading all its way. Remember that forex market is participated by traders all over the world. Thus it has to be available for trading in normal business hours at local time for every country.
Forex 24 hours trading is very different from, say, 24 hours trading on the index futures.
Take emini S&P as an example, its underlying index S&P 500 is updated in North America during business hours in North America only. Thus afterhours trading on the emini S&P has no reference to which how the actual S&P 500 will do until the next business day. That factor alone makes afterhours trading of the emini S&P very illiquid.
Comparing that to forex market, every trading day there are three (3) distinguish time period that forex market gets very busy. That is, the start of the business day for Asia, Europe, and North America. It makes perfect sense because most traders need to sleep, and most of the participants in the forex market of each region will start their trading day at their normal business hours in their home countries.
In short, the 24 hours trading in forex increases its overnight risk significantly comparing to stock and index futures. That is a good reason why protective stop orders should be used if you are carrying your position overnight in forex, because while you are sleeping someone else is actively trading it in other parts of the world.
Daily Bars are not Really Daily Bars
If you can read multiple languages, then you can easily compare the financial section of newspapers from various countries, either online or getting physical copies. One thing you will notice is that they all quote a different set of closing prices for the forex markets. Some newspapers are more knowledgeable and printed the exact time that the forex prices were recorded.
Since every country quotes a different set of closing prices for the day, there is no real meaning to what we known as daily price bars. The net change from the previous trading day has a different meaning for traders using North America time and those who trades using Greenwich Mean Time (GMT).
When the daily price bars are not consistent, that affects the validity of the so-called pivot points (originally known as floor trader support/resistances levels) indicators. different traders will have a different set of price levels calculated, depending on their location, and time range they use for the construction of their daily prices.
Another type of technical tools that are affected by the inconsistencies of daily bars are long term trend lines. As oppose to using the trend lines in a very precise manner, you may have to relax the condition and/or wait for confirmation from the prices before you can tell if your trend lines are holding up or being violated in real-time.
Historical Data Consistency
Forex data does not include actual traded transaction data (forex is a dealer network based market, not a centralized auction market), you only get the latest bid and ask update information. Historical charts are constructed from the bid and ask streams. Most data vendors choose to use the bid stream as the reference price for the construction of historical data. Some choose to use the ask stream instead. And, for some forex brokerages, they take the mid-point value between the bid and ask prices for the construction of the historical data series.
As I mentioned in the first article, as there is no trade information, there is no volume analysis. Tick analysis is also not possible.
For system developers, you have to know clearly how your historical data is constructed. If it is bid stream based, all the buy side market orders will be off by at least the spread. For example, your brokerage offers 3 pip spread for trading EUR/USD. Then if you backtest your system against the historical data provided by this data vendor, your historical market buy orders will be filled at the bid price, not the ask price.
If the historical data is constructed from historical ask stream, the inverse is true – the historical sell orders will be off by at least the spread size.
One way to compensate for this error is to enable limited worst case analysis in NeoTicker and set the number of ticks in slippage to the size of the spread. Then you will get a more accurate account of the performance of your system historically.
Longer Consolidation Period
During periods where most market participants are not working, you will find the forex market becomes extremely slow. For example, after general US market close at around 4 PM Eastern Time, there is quite a number of hours before Asia financial markets open. During this period of time, the market actions will be very slow and affecting many oscillator based indicators.
Most momentum type or oscillator type indicators do not work well when the underlying price series has a prolonged period of time staying in a tight range.
Traders can stop trading during these slow periods to avoid the noisy signals, or, using higher time frames like 2-hour bar or 3-hour bars to get better overall reading of the current market condition.
You can also use techniques to transform the forex data into something that is independent from time, like Price Range superposition, point and figure charting, etc.
I think I have covered enough differences for traders who have trading experience with stocks or index futures to get started in studying the forex market.