Back to Community
Another VIX trading algorithm using the new ETF - VMIN

VMIN is a new volatility ETF that uses weekly futures than the monthly one. It's too short a time period to say anything about this algorithm, but it does handle multiple dips last year like a champ.

Clone Algorithm
62
Loading...
Backtest from to with initial capital
Total Returns
--
Alpha
--
Beta
--
Sharpe
--
Sortino
--
Max Drawdown
--
Benchmark Returns
--
Volatility
--
Returns 1 Month 3 Month 6 Month 12 Month
Alpha 1 Month 3 Month 6 Month 12 Month
Beta 1 Month 3 Month 6 Month 12 Month
Sharpe 1 Month 3 Month 6 Month 12 Month
Sortino 1 Month 3 Month 6 Month 12 Month
Volatility 1 Month 3 Month 6 Month 12 Month
Max Drawdown 1 Month 3 Month 6 Month 12 Month
# Backtest ID: 58a8e99398e74e5e04505ceb
There was a runtime error.
3 responses

You may be seeing an anomaly because of the low trading volume of VMIN and the interplay with order rounding because of the low $1000 starting value in your backtest.

Here's the same algorithm but with $10,000 starting capital and the following line of code added (to minimize effects from orders not being filled).

    set_slippage(slippage.FixedSlippage(spread=0.0))

Note the 39% volatility and 20% drawdown. Much higher than the backtest using $1000 starting capital. It does have a 155% return though.

Clone Algorithm
63
Loading...
Backtest from to with initial capital
Total Returns
--
Alpha
--
Beta
--
Sharpe
--
Sortino
--
Max Drawdown
--
Benchmark Returns
--
Volatility
--
Returns 1 Month 3 Month 6 Month 12 Month
Alpha 1 Month 3 Month 6 Month 12 Month
Beta 1 Month 3 Month 6 Month 12 Month
Sharpe 1 Month 3 Month 6 Month 12 Month
Sortino 1 Month 3 Month 6 Month 12 Month
Volatility 1 Month 3 Month 6 Month 12 Month
Max Drawdown 1 Month 3 Month 6 Month 12 Month
# Backtest ID: 58a9bc0ccfc31e5deee1cc9a
There was a runtime error.

Here's the same backtest but trading XIV instead of VMIN.

This has 30% volatility and 16% drawdown (vs 39% and 20% using VMIN). It does have lower returns though... 129%.

Not making any statement about which ETF to trade but simply highlighting that maybe the original backtest may not have been an accurate comparison. It would be interesting to analyze and understand how the unfilled orders in the first backtest ($1000 starting cash and default slippage) worked to cut the drawdown in half. Maybe incorporate that anomaly as part of the algorithm code?

Clone Algorithm
63
Loading...
Backtest from to with initial capital
Total Returns
--
Alpha
--
Beta
--
Sharpe
--
Sortino
--
Max Drawdown
--
Benchmark Returns
--
Volatility
--
Returns 1 Month 3 Month 6 Month 12 Month
Alpha 1 Month 3 Month 6 Month 12 Month
Beta 1 Month 3 Month 6 Month 12 Month
Sharpe 1 Month 3 Month 6 Month 12 Month
Sortino 1 Month 3 Month 6 Month 12 Month
Volatility 1 Month 3 Month 6 Month 12 Month
Max Drawdown 1 Month 3 Month 6 Month 12 Month
# Backtest ID: 58a9bba02169cc5de856321e
There was a runtime error.

Exception: Problem reaching https://www.quandl.com/api/v3/datasets/YAHOO/INDEX_VIX.csv
Anyone have a solution?