Summary:

• The AiLA allocation process is demonstrated using a research portfolio comprising five assets.
• The portfolio can at any time have long or short asset allocations, following the AiLA index methodology [Method].
  [Method]: Products - Methodology , https://www.ailaindices.com/strategies.php
• The case study attempts to address the questions,
  • What drives the long/short allocation for a certain asset at a particular time?
  • What drives the exit opportunity for an invested position?
  • What determines how much to be allocated to a given asset?

Asset allocation decision

• ML model used to identify allocation opportunities at a per asset level, e.g. WTI March and December contracts are modelled separately.
• Generates a long or short allocation signal, with entry and exit time depending on the expected risk/reward for the individual asset.

Entry decision

• Asset allocation entry is determined by ML model predicting a favorable risk/reward opportunity for a long or short allocation.
• The ML model prediction is based on features generated from fundamental as well as market price data, e.g. including data such as
  • Seasonal: month, year, week / business day of the year.
  • Price: open/close/high/low for different contracts and markets.
  • Fundamental: commodity production/import/stocks, macro indices, FX rates, CFTC commitment of traders reports.

Exit decision

• An allocation will be kept until one of the following exit criteria is met.
• 1. Expects little opportunity to be invested further.
• 2. Hitting the risk target since start of investment.
• 3. Achieves the reward target since start of investment.
• 4. Receiving an opposing opportunity from the model, using features data.
• Above, only exit criteria (4) depends on the features data

Example: June Gold (CME) - Long allocation

2022-04-13: Entry decision

• Favorable risk/reward predicted based on 38 features

(* CFTC CoT NCNC = CFTC Commitments of Traders, Non-Commercial Net Contracts)

2022-04-21: Exit decision

• Hitting risk target since start of investment

Portfolio allocation weights

• Portfolio weights are expressed as percentage of total portfolio allocation.
• Determined by portfolio methodology based on upstream asset level allocation opportunity signals.
• Example portfolio: July Bean, June Copper, June Gold, Cal TTF and June WTI.
• Portfolio target: equal 20% weight per asset, however, allow asset weights up to 100%.

Opportunistic allocation

• 2022-03-23: allocation opportunities for Gold, Copper and Soybean, weights increased above 20% resulting in system allocating 94.1% of capital.
• 2022-05-16: TTF only asset with allocation opportunity, 20% weight prevented by liquidity constraints resulting in system allocating 6% of capital.
• 2022-05-17: TTF opportunity ends, resulting in system decision to exit 6% long TTF weight.

Varying number of opportunities

• Periods with large (small) number of active assets tend to result in smaller (larger) individual weights.