Evaluation of momentum and trend following strategies applied to cryptocurrencies looks promising

Discussion in 'Crypto Assets' started by TauLepton, Dec 29, 2017.

  1. I've recently been interested in the prospect of applying a trend following system to cryptocurrencies. We have consistently seen incredibly large increases in the values of cyrpto's over the past few years which leads me to belive they are the perfect candidate for a trend following system.

    The university of Zurich recently released a paper in which they evaluated the performance of momentum and trend following strategies applied to cryptocurrencies. The results look very promising, to summarise:
    "The result is a 42.02% annualized return and a Sharpe ratio of 1.4843 for the time series portfolio. The crosssectional portfolio achieves an annualized return of 56.94% and a Sharpe ratio of 1.6793."

    Is anyone on ET using such a system to trade cryptos?

    The paper can be found here: https://pd.zhaw.ch/publikation/upload/213002.pdf
     
    Gambit likes this.
  2. Cuddles

    Cuddles

    I wonder how their control stacked up.
     
  3. Sorry I don't understand what you mean, care to elaborate?
     
  4. can the entries/exits hold in real trading? liquidity & wide spreads are big issues in btc trading. in an instrument where cross exchange arbitrage is inefficient, I expect slippage & no fills when trading it

    momentum & trend strategies should work on ANY instrument. but context on when to use the appropriate strategy is what makes a successful trader
     
    cvds16 likes this.
  5. What`s being traded?Futures, CFDs, etc?For the CFDs spreads are too high for the momentum trading.
     
  6. Some of the larger exchanges have more than sufficient liquidity to handle the capacity at which I would be looking to trade. The larger exchanges also have tighter spreads. Correct me if I'm wrong but I believe volatility normalised spreads is what is primarily important for trend following and I suspect these will be small for cryptos but I would need to do some backtests to quantify this.

    Crypto/USD pairs are being traded.
     
  7. Fileiro

    Fileiro

    I'll venture a guess that none of the researchers behind this study are crypto-billionaires which would answer the question whether the strategy works in the real world as well..
     
    cvds16 likes this.
  8. Yeah, the results`d looked very promising until it went on CME.
     
    cvds16 likes this.
  9. If you take a look at one of the authors, Joerg Osterrieder, profile you will see that he has a very well decorated history both professionally and academically. https://www.zhaw.ch/en/about-us/person/oste/

    Irrespective of the authors credibility, the time series doesn't lie. There has been consistent increases in price on many cryptocurrencies which leads me to believe they would lend themselves very well to a trend following system.

    Why do you think bitcoins listing on CME changed things? Are you suggesting bitcoin no longer trends? In any case there is a large number of cryptocurrency candidates which are not listed on CME.
     
  10. Bitcoin's intraday chart is old school TA as good as it gets!!
     
    #10     Dec 30, 2017
    Visaria and cvds16 like this.