TRADERTALK
when short-term trends are changing. Tis means that when banks take the other side of our trades they are still able to profitably cover their exposure to us. Tis makes our strategies very scalable – in the world’s deepest markets. We have developed our own execution strategies and we have our external IT partners refining some new execution algorithm strategies for us.
How do you go about back-testing your proprietary currency trading systems and signals to confirm that your strategies will perform as required?
We have been testing models continuously for five years. Te way things develop is that Riadh or I toss out an idea about a possible model and then we think about how that would work and whether it is complementary to our existing model sets. We then begin building it out and thinking about what time period may be appropriate. Most of our model sets spend the majority of their time looking at 1-minute to 240-minute time “slices.” We analyse the change of market conditions within that time period and compare it other time periods. Once we are comfortable that a new model could be an interesting portfolio addition we begin a series of tests.
It is not unusual that due to all the filters that we usually deploy, a single test using only one year of tick data could take a week of 24- hour computing time to complete. We are analysing all the iterations of every bid/offer spread, 24 hours a day for approximately 250 market days a year. To test several years of data and then study the results and make any necessary changes may take several months. What many people do not necessarily appreciate about our testing methodology is that for us to test a single calendar year of intraday data, on a single model, on a single currency pair, would equate to a medium-term trend follower testing over 86,000 years of a daily data series. We prefer the testing process to be, as close as is possible of course, to “real world” conditions. I have long since given up printing the test results as we filled one room in Switzerland with binders of test ouputs. We increased our growing collection
190 | april 2012 e-FOREX
of multi-terrabyte external drives to store cheaply and efficiently what has become a vast array of data storage, the likes of which if printed on paper would now take a large warehouse to store.
When we are happy with test results (which is a about 5% of the time), we initiate the model into a separate live Demo server for observation. Tis typically runs for a month or two on live data, but without live execution, in order to make sure that the observed results are in line with expectations. If we are pleased with the results then we initiate a test with proprietary capital in a separate account for several months in order to make sure the actual results generated are as to be expected. You can begin to see why it was over three years before we first launched our incubation account! We don’t know too many firms who have enough conviction in their models that they are prepared to fund three years of research without any revenue.
What trading platforms do you use and what factors influenced your choice?
We spent the first couple of years giving out projects to external IT and algo consultants. Unfortunately we found them to be expensive, arrogant, slow and poor project managers. We then brought our programming resourcing in-house and we used MATLAB and C++. We found this process was also slow and liable to error between what the programmers crafted and what we wanted. Te final outcome is that after experimenting with a number of external and internal solutions we found that MT4 could do about 80% of what we needed and it was a very easy and flexible system for us to update and modify. We commissioned the additional add-ons that we wanted and then we experimented with a couple of bridge providers before settling on PrimeXM.
How did you go about building your trading desk IT infrastructure and was the trading software and connectivity technology provided by third party vendors?
We originally started internally, then we bought a server to plug into a data center and then we finally went for a secure Cloud Computing solution. We still use our own server for testing and we use servers in both London and NY to demo test and to trade. We want our trading servers to be close to the bank servers. We have an IT service firm in Switzerland that we have used for many years to
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102 |
Page 103 |
Page 104 |
Page 105 |
Page 106 |
Page 107 |
Page 108 |
Page 109 |
Page 110 |
Page 111 |
Page 112 |
Page 113 |
Page 114 |
Page 115 |
Page 116 |
Page 117 |
Page 118 |
Page 119 |
Page 120 |
Page 121 |
Page 122 |
Page 123 |
Page 124 |
Page 125 |
Page 126 |
Page 127 |
Page 128 |
Page 129 |
Page 130 |
Page 131 |
Page 132 |
Page 133 |
Page 134 |
Page 135 |
Page 136 |
Page 137 |
Page 138 |
Page 139 |
Page 140 |
Page 141 |
Page 142 |
Page 143 |
Page 144 |
Page 145 |
Page 146 |
Page 147 |
Page 148 |
Page 149 |
Page 150 |
Page 151 |
Page 152 |
Page 153 |
Page 154 |
Page 155 |
Page 156 |
Page 157 |
Page 158 |
Page 159 |
Page 160 |
Page 161 |
Page 162 |
Page 163 |
Page 164 |
Page 165 |
Page 166 |
Page 167 |
Page 168 |
Page 169 |
Page 170 |
Page 171 |
Page 172 |
Page 173 |
Page 174 |
Page 175 |
Page 176 |
Page 177 |
Page 178 |
Page 179 |
Page 180 |
Page 181 |
Page 182 |
Page 183 |
Page 184 |
Page 185 |
Page 186 |
Page 187 |
Page 188 |
Page 189 |
Page 190 |
Page 191 |
Page 192 |
Page 193 |
Page 194 |
Page 195 |
Page 196