Haven't got time to read through your threads yet. However, first at all. I think perhaps you would need to do some sort of of ABC analysis https://en.wikipedia.org/wiki/ABC_analysis and/or PCA analysis https://en.wikipedia.org/wiki/Principal_component_analysis to determine and let us know your main focus of income performance. A: Trading income, such as your trading edge and expected returns (for this category alone), money management logic, risk protection/ hedging, etc. Also the benefits of your full automation systems. My guess is ETers would be more interested in this category of topics. B: Long Term Investment income, such as international stocks, overseas properties, etc. Whether and how you perform any currency overlay for this category of incomes and related profit and loss performance analysis could be only covered/ disclosed/ discussed later, separately with different threads. C: Non-financial income, such as book publishing for financial programming, risk consultancy to financial institutions, etc. These could be minor issues to most ETers (in terms of discussion posts), I guess. Unless you would like to disclose your trading Edge in details here. Just some thoughts!
Well usually I stick to 'A', except when responding to a question as earlier in the thread. But I'll bear this in mind. Generally if there are specific topics people would like to hear about, then I'm happy to hear what they are. GAT
After briefly reading the first 2 pages, I can see there could be 3 questions. 1. Systemic risk. I think, just unsure whether adding more instruments (like 40 - Pros and Cons) would reduce the portfolio's systemic risk. It could be. I could be most likely not (due to such as some overseas exchanges still close/illiquid during crash). 2. SAR strategy. My guess is using robot for SAR strategy (without your personal ongoing attention) could possibly sometimes encounter huge risk/loss if volatility is drastic. (perhaps back-testing with intraday data would be required/ necessary) 3. Hedge. Just unclear how you perform hedging for the 'A' trading portfolio (while maintaining good returns). ( I assume All your hedge is just for 'B' Investment. If not, then even more confusing!)
It's Stop And Reverse. I suppose your futures positions are always-in-the-market . Now I'm afraid maybe I got my misunderstanding .
1. I guess the definition of systematic risk is that it won't be reduced through diversification. Will it be increased? Are we talking about market risk, or other risks (like market closure, which you mention). With market risk I think the biggest danger is you add diversification and then ramp up gearing in line with perceived correlation and the fall in vol. I have measures to prevent that; limits on the 'ramping up multiplier' and a check on the total size of my positions assuming all correlations 'go to one'. (http://www.elitetrader.com/et/index...ed-futures-trading.289589/page-5#post-4092502). With non market risk; well I suppose spreading across multiple exchanges / clearers is probably safer than not. I'm UK based so nearly all the markets I trade are 'foreign'. By the way one thing that would worry about cross country trading is where it's done on a relative value basis. Would you want to be long UK / short US after september 11th happened? No way; you'd be getting margin calls on your UK position whilst receiving no benefit from your US short whilst the market was closed. 2. I'm unclear here if you are talking about a) a sharp jump / gap in prices b) a sharp move in prices over the day, which for some reason the robot (a term, by the way I hate) doesn't trade out of c) the robot doing something crazy on a day when prices seesaw up and down wildily. With (a) I guess humans and robots are both equally vulnerable (although a low latency robot - not what I'm running - would react faster). However you're right that daily market data will give a rosy picture of peak losses compared to intra day. With (b), perhaps in the event of system failure this is a problem; but this needs to be balanced against the benefit of not having an emotional human debating with itself whether to cut or not; it would just cut. I think you're talking about (c). Few points to make here: One is I trade relatively slowly (average holding period of perhaps a month), so I don't react much to intraday moves. Suppose I was long Emini and the market dropped 300 points; then reversed and rose 300 points, and then dropped 300 points. On the first drop I would have cut my position massively; partly because the trend has started to fade, mainly because volatility has spiked. On the move up I wouldn't do anything. Volatility hasn't fallen, if anything it's gone up. On the next down move again I wouldn't do anything. Two, I have controls in place to prevent over trading. The crudest is a limit on the number of contracts I will trade per day. I also have a 'lifetime' limit which I would set before a holiday where I can't monitor my trading. Here is an excerpt Code: run@bilbo ~/workspace/systematic_engine/sysdiag/scripts $ . displaylimits LIVE EDOLLAR 2015-06-18 12:07:42 Max pos:40 Current pos 5 Max day:10 Done:2 Max all:1000 Done:259 SP500 2015-06-18 06:05:41 Max pos:6 Current pos 0 Max day:2 Done:0 Max all:1000 Done:61 For S&P after I've traded two contracts I'll stop. Right now with a max strength signal I'd own 2 or at most 3 contracts (although the maximum position I'd allow is 6 contracts as shown). A trade of two contracts in a day is plenty (or two trades of one contract each). Were I to go on holiday I'd set the lifetime limit (max all) maybe to 10 contracts for a 2 week holiday. Notice the limits for Eurodollar, a lower risk contract, are higher. 3. Hedging. It's a very simple hedge of two equity portfolios with two equity index futures (eurostoxx and FTSE); roughly delta neutral (or as close I can get given the size of the futures contracts). GAT
OK, thanks for your response. Look forward to viewing your 'A' trading annual performance in the future.
Sporadic update (been about 2 weeks since the last one) So down ~20K, or 5% if you prefer. Drawdown is -17.5%; slightly better than the worst level I've seen. As the picture shows the last few weeks I've just been "bobbling" around. I've said it before the trick with trend following is not to lose too much when you're losing; and win more when you gain. My long term return expectation is perhaps 18%. Last fiscal year I made 57% (or would have with constant risk - I actually did much better but that wasn't down to my trading system). This fiscal year I'm down 10%. Even if I lose another 10% this year I'd still be ahead of expectation. Big losses 4.K Gas 3.1K JPYUSD 2.5K BTP 2K Gold 2.0K Crude 1.6K MXPUSD Big gains 3.3K Palladium 2.4K NZDUSD 1.5K copper 1.4K corn 1.2K Platinum Risk is pretty much unchanged; i.e. still low. No big trends to catch on to right now. Current positions Code: code contractid positions Lock WrongContract InFwdNotRoll 16 BTP 201509 -1 False False False 12 COPPER 201512 -1 False False False 13 CORN 201512 -5 False False False 15 EDOLLAR 201812 3 False False False 18 EDOLLAR 201809 2 False False False 8 EUROSTX 201509 -18 False False False (hedge) 17 GAS_US 201508 -1 False False False 10 JPY 201509 -1 False False False 14 KOSPI 201509 -1 False False False 4 KR3 201509 5 False False False 6 MXP 201509 -5 False False False 11 NZD 201509 -2 False False False 19 PALLAD 201509 -2 False False False 1 PLAT 201507 -3 False False False 9 SOYBEAN 201511 1 False False False 5 US2 201509 3 False False False 3 US5 201509 1 False False False 0 V2X 201508 3 False False False 2 VIX 201508 -4 False False False 7 WHEAT 201512 -1 False False False Of interest is that I've closed all my australian markets (AUS3, AUS10 bonds, AUSSTIR and ASX) due to an increase in data feed costs. I probably have too many markets trading (quite a few positions a bit small), so I don't mind a little pruning. Also closed my FTSE hedge of 2 contracts and replaced with an increased Eurostoxx hedge. Trades Code: code contractid filled_datetime filledtrade filledprice 4003 AUD 201506 2015-06-10 06:43:28 1 0.767900 4006 AUD 201509 2015-06-10 06:43:28 -1 0.764000 4201 AUD 201509 2015-06-18 08:47:52 1 0.772800 4114 AUS10 201506 2015-06-12 06:50:07 -1 96.987500 4117 AUS10 201509 2015-06-12 06:50:19 1 96.940000 4186 AUS10 201509 2015-06-18 06:54:17 -1 97.110000 4120 AUS3 201506 2015-06-12 06:51:13 -2 97.910000 4123 AUS3 201509 2015-06-12 06:51:24 2 97.880000 4189 AUS3 201509 2015-06-18 06:55:11 -2 98.070000 4192 AUSSTIR 201606 2015-06-18 06:57:12 -7 97.970000 3970 BTP 201509 2015-06-08 07:32:04 -1 130.590000 4081 BTP 201509 2015-06-11 13:58:12 1 132.110000 4147 BTP 201509 2015-06-16 07:35:08 -1 127.840000 4198 BTP 201509 2015-06-18 07:31:21 1 129.470000 4216 BTP 201509 2015-06-18 09:46:42 -1 129.720000 3964 COPPER 201512 2015-06-05 15:29:21 -1 2.717000 4087 CORN 201512 2015-06-11 14:30:00 1 373.000000 4039 CRUDE_W 201512 2015-06-10 12:16:38 1 62.900000 4225 EDOLLAR 201812 2015-06-18 12:07:23 2 97.535000 4162 EUROSTX 201506 2015-06-16 15:32:19 8 3442.000000 4165 EUROSTX 201509 2015-06-16 15:32:19 -8 3437.000000 4255 EUROSTX 201509 2015-06-19 13:53:52 -10 3478.000000 4210 FTSE 201506 2015-06-18 09:36:09 2 6671.000000 4213 FTSE 201509 2015-06-18 09:37:04 -2 6617.500000 4252 FTSE 201509 2015-06-19 13:46:46 2 6677.500000 3976 GAS_US 201508 2015-06-08 16:45:52 1 2.708000 3997 GAS_US 201508 2015-06-09 14:23:51 1 2.841000 4078 GAS_US 201508 2015-06-11 12:11:11 1 2.903000 4033 GBP 201506 2015-06-10 07:47:00 1 1.537300 4036 GBP 201509 2015-06-10 07:47:00 -1 1.536200 4063 GBP 201509 2015-06-11 01:42:42 1 1.549100 3973 GOLD 201508 2015-06-08 12:08:23 -1 1173.700000 4075 GOLD 201508 2015-06-11 11:59:07 1 1178.400000 4249 GOLD 201508 2015-06-19 12:05:23 1 1200.200000 3979 JPY 201506 2015-06-08 19:26:14 1 0.008025 4000 JPY 201506 2015-06-10 06:13:13 1 0.008127 4009 JPY 201506 2015-06-10 06:46:46 2 0.008041 4012 JPY 201509 2015-06-10 06:46:46 -2 0.008049 4183 JPY 201509 2015-06-18 01:51:06 1 0.008113 3967 KOSPI 201506 2015-06-08 01:27:52 -1 253.400000 3982 KOSPI 201509 2015-06-09 01:01:07 -1 254.750000 3985 KOSPI 201509 2015-06-09 01:03:07 -1 254.750000 3988 KOSPI 201506 2015-06-09 01:04:51 1 253.900000 3991 KOSPI 201509 2015-06-09 01:08:27 1 254.950000 4138 KR3 201506 2015-06-16 01:01:23 -3 109.290000 4141 KR3 201509 2015-06-16 01:06:11 2 109.150000 4171 KR3 201509 2015-06-17 01:02:00 1 109.100000 4237 KR3 201509 2015-06-19 01:04:49 2 109.090000 4027 MXP 201506 2015-06-10 06:52:27 5 0.064240 4030 MXP 201509 2015-06-10 06:52:27 -5 0.063830 4015 NZD 201506 2015-06-10 06:47:35 1 0.714000 4018 NZD 201509 2015-06-10 06:47:35 -1 0.708000 4021 NZD 201506 2015-06-10 06:49:59 1 0.714000 4024 NZD 201509 2015-06-10 06:49:59 -1 0.708000 4072 NZD 201509 2015-06-11 01:58:41 1 0.696900 4144 NZD 201509 2015-06-16 01:45:57 -1 0.693700 4159 PALLAD 201509 2015-06-16 15:31:03 -1 736.250000 4135 PLAT 201507 2015-06-15 16:28:28 -1 1087.200000 3994 SOYBEAN 201511 2015-06-09 12:14:58 1 922.250000 4126 SOYBEAN 201511 2015-06-15 12:19:59 -1 898.750000 4168 SOYBEAN 201511 2015-06-16 18:15:58 1 922.500000 4228 SOYBEAN 201511 2015-06-18 12:05:40 1 935.750000 4234 SOYBEAN 201511 2015-06-18 16:36:39 1 940.250000 4231 US2 201509 2015-06-18 14:05:52 -1 109.367188 4084 US5 201509 2015-06-11 13:59:25 -1 118.359375 4258 US5 201509 2015-06-19 14:36:31 1 119.328125 4150 V2X 201508 2015-06-16 08:23:21 1 22.600000 4240 V2X 201508 2015-06-19 08:10:32 1 22.650000 4207 VIX 201507 2015-06-18 08:49:54 4 15.600000 4219 VIX 201508 2015-06-18 10:03:35 -4 16.400000 As you'd expect for this time of year a LOT of rolls; as well as the aforementioned closing and rehedging trades. Slippage was £281 vs an expectation of £402, or if you like my execution algo saved me £120.
Correction: My long term return expectation is perhaps 16%. Last fiscal year I made 57% (or would have with constant risk - I actually did much better but that wasn't down to my trading system). This fiscal year I'm down 10%. Even if I lose another 14% this year I'd still be ahead of expectation. Addition: My worst drawdown in back test was 33%. If I lost another 14% I'd be at a 31.5% drawdown. So this is at the bottom end of my expectations. (Percentages are as total of maximum capital at risk so I can just add and subtract them rather than using geometric mean)
BTW: Are you trading any of the following: XINA50, STW, HSI, HHI.CH, NIFTY, TOPX? I've been having fun with XINA50, and there is a trend there (although I just trade it in a daytrading fashion).