Sunday, 28 August 2016
Last updated 1 day ago
Jul 10 2014 | 10:39am ET
You may call K1T Capital many things—systematic, quant-based, hedge fund—but don't call it a trend follower:
“We exploit periodic cyclicality in financial data series,” CIO Ben Heaton told FINalternatives. “Our trade positioning is counter-cyclical making us an MR (mean reversion) system which is rare amongst algorithmic approaches, the vast majority of which are trend followers. This is a strength for us as it means our returns have low correlation to the crowd.”
Added CEO Simon Wajcenberg: “K1T’s edge lies in using scientific research to capture patterns in market behavior that, while hard to discover, do exist and can provide an advantage for those able to exploit them.”
The career paths that led both Wajcenberg and Heaton to founding a quant hedge fund were unorthodox, to say the least.
“My background is entrepreneurial,” Wajcenberg told FINalternatives. “I have founded, managed and sold numerous businesses. In the last 15 years I have worked in the online environment and more recently in mobile media.”
Three of Wajcenberg's former ventures—the online broker Stocknet.co.uk; the real-time price information provider Freequotes.co.uk; and the TMN group, an online direct marketing company—were launched with Heaton.
Post-TMN, Wajcenberg remained in online/mobile advertising while Heaton studied medicine and became a doctor, spending three years working for the UK's National Health System.
But Heaton, who had been a quant analyst and Bund trader at Paribas prior to teaming with Wajcenberg, didn't entirely give up on the world of finance:
“Ben has been developing the algo that is the backbone of our trading system for over 15 years,” said Wajcenberg. “Around 12 months ago, it became clear that he had a profitable solution and trading system.”
Heaton invited Wajcenberg to bring his sales and marketing expertise to K1T Capital, which launched in 2013 with Peter Coveney, another TMN vet, as CFO.
“I was and am excited about the prospects for K1T Capital and felt we made a strong team and have a robust and profitable model. I therefore agreed to join Ben and launch our fund. The K1T Capital Quant Master Fund 1 SP.”
The first 18 months of K1T's existence were spent developing the trading system that went live with internal capital in April of 2014.
Back-tests indicated the strategy would have returned 20% over the six years from 2008 to 2013. Said Heaton:
“In the past, researchers have overlooked the possibility of periodic cyclicality as being too simplistic and have sought more complex solutions. We allowed a Big Data approach to tell us what worked and what didn't—and didn't argue with the results.”
K1T's automated trading system is actually eight separate systems (one of which is still in development), each of which has slightly different parameters. TS1, for example, is a 100% hedged strategy that takes long and short positions in S&P 500 stocks; TS5 takes long-only positions in global stock market indices using the iShares global stocks ETF series; TS8 is an options strategy.
“They all have the same algorithm,” said Heaton, “but different parameter optimizations. For instance, U.S. S&P 500 stocks have a cycle [with] an average wavelength between 10 and 11 trading days, while US listed gold stocks have a wavelength between 15 and 20 trading days.”
The core algorithm is fixed, but each month all the parameters for the algorithm are optimized on the previous 10 years' closing price data for each stock.
“We use a Big Data approach to analyze all possible combinations of parameters and generate the Sharpe Ratio for each parameter set,” said Heaton. “Only those parameter sets with historical Sharpe ratios that exceed a threshold are accepted for generating new signals in the month ahead. At the end of the month, the whole process is repeated. It is a fully automated, autonomous process of parameter optimization.”
Each trading system has a weighting within the fund: global ETFs, for example, count for 25%.
Other fund features are low leverage—the maximum leverage used over the six-year back-testing period was 150%; liquidity—it invests only in highly liquid markets; and ease of redemption—there is no lock-in or redemption period, funds may be withdrawn with 10 days' notice prior to the end-of-month redemption date with no penalty.
The firm has also recently bolstered its management team with the addition of three industry veterans to its advisory board: Craig Morgan, a 15-year Morgan Stanley vet; Simon McDonald, whose resume includes TD Securities, Lehman Brothers, Commerzbank and Merrill Lynch; and James Hogbin, the former COO Systematic Asset Management, an RBS systematic quant fund.
K1T's minimum investment is $100,000 and its target investors are high-net-worth individuals and family offices.