Fortuitous timing on your decision with spot VIX moving from 17 to 37 today! Hope everyone did well, or at least not too badly, after last night.
Honestly, I've never been so lucky in my life. Literally the last hour. Very, very fortunate indeed. -19% from HWM. System is cutting back, but still strongly long equities.
Sometimes being lucky is better than being good I've mulled various 'proper' ways of dealing with the skew/ kurtosis of the VIX in considering position sizes; keeping a relatively small allocation certainly helps, as does applying a vol floor, but I also wonder whether one should use some kind of percentile value of historic volatility rather than the standard deviation. Any such method will mean you undershoot your target risk on VIX / V2X most of the time however. Perhaps it is better to just give it a smaller risk allocation, and deal with the downside when it happens. You do after all get paid for being short vol, in the long term. I'm currently at -12.3% from HWM FWIW Here is last nights risk report, would have budged slightly today (in particular the VIX / V2X positions - had I been holding them - would have reduced sharply. I have a manual price protection which filters out larger price movements until manually checked - this normally goes off once a year, but went off every hour for vol as the moves were around 10 - 15 standard deviations). It's flat equities, short bonds; and long some commodity stuff which unfortunately are still correlated with equities. Code: code multisignal expected_annual_risk 16 V2X -16.4 0 [override] 17 VIX -20.8 0 [override] 22 NASDAQ 0.0 0 2 LIVECOW -0.0 12 21 SMI 0.0 0 19 AEX 0.0 0 8 BTP 0.0 0 11 SHATZ 0.0 0 20 CAC 0.0 0 23 SP500 0.0 0 37 EUROSTX 0.0 0 [actually short in equity hedge] 18 KOSPI 0.0 0 14 US20 -0.9 606 13 US2 -1.1 750 6 KR3 -2.0 1380 15 US5 -3.1 2191 7 BOBL -1.5 1050 12 US10 -3.3 2323 5 KR10 -6.0 4199 9 BUND -5.8 4049 10 OAT -7.7 5425 36 EDOLLAR -16.6 11613 0 CORN -13.9 9709 4 WHEAT -16.5 11578 1 LEANHOG 1.1 801 24 AUD 8.3 5824 27 JPY 4.6 3214 30 COPPER 7.4 5180 25 EUR 8.7 6122 29 NZD 10.3 7215 31 GOLD 2.0 1418 33 PLAT 10.7 7466 26 GBP 9.2 6432 35 GAS_US 9.1 6354 3 SOYBEAN 12.4 8683 28 MXP 19.8 13893 32 PALLAD 19.6 13712 34 CRUDE_W 19.8 13861 GAT
Did you manually take off equities or did the system do that? The reason I ask is because my system was staying very long for equities, and finally I had enough & shut them all myself. re: Vix/Vstoxx, I concluded it wasn't fair to lump these in with trend following, as the return distribution isn't what you'd expect (trend + gaussian noise).
I wonder if anyone has tested a "volatility filter" within their system? Meaning that for market shocks like we are currently moving through the system could be programmed to ignore position changes when market volatility is over a certain level. The rational being that volatility begets more volatility and we can expect some recovery. I may be talking my own book
Any idea which rule dominated that decision? My own rule weights are: rule_weights = {'carry': 1.0761308295764749, 'ewmac16': 0.94565802638498464, 'ewmac32': 0.97404151565516006, 'ewmac64': 1.037272611390585, 'ewmac8': 0.96689706277925536, }
I don't have the diagnostics to easily check, but my weights for SP500 are: Code: SP500 breakout10 0 breakout20 0.143 breakout40 0.1144 breakout80 0.0286 breakout160 0.02145 breakout320 0.05005 mrinasset80 0.02 mrinasset160 0.08 carry10 0.08 carry30 0.06 carry60 0.04 carry125 0.02 assettrend2 0 assettrend4 0.039 assettrend8 0.0312 assettrend16 0.0078 assettrend32 0.00585 assettrend64 0.01365 normmom2 0 normmom4 0.039 normmom8 0.0312 normmom16 0.0078 normmom32 0.00585 normmom64 0.01365 momentum2 0 momentum4 0.039 momentum8 0.0312 momentum16 0.0078 momentum32 0.00585 momentum64 0.01365 relcarry 0.05
From the FT today: "Trend-following quantitative hedge funds were among the biggest victims of the current market turmoil, wiping out much of their 2018 gains in two torrid days of trading and undercutting their status as risk-mitigation tools. Surging global share prices during January lured in more money from computer-driven hedge funds that ride market momentum, but the sudden eruption of volatility triggered abrupt losses for many of the industry’s biggest players. “The rally had accelerated, and we kind of knew an accident would happen, but it’s never pretty when it does,” said David Harding, the head of Winton Capital, one of the biggest such funds in the industry. “We’ve had a couple of painfully bad days . . . This will go down as one of the weeks one remembers.” Mr Harding said Winton’s funds were down 3 per cent to 4 per cent over Monday and Tuesday, their worst short-term performance in five years, according to the hedge fund manager. Yet Winton’s funds use relatively little leverage and the pain is likely to have been more intense elsewhere. "
Trend following strategies tend to have negative skew on the daily timescale as mentioned in this thread many pages earlier. On the monthly scale the skew is positive however.