The “Naïve” class of portfolios defines the weights heuristically, based on human experience and feelings.
Take for example the case of portfolio where the weights are set to be proportional to the inverse of the portfolio components volatilities experienced during a predefined historical period. This portfolio construction is motivated by the market wisdom that the exposure to an asset with high volatility should be limited. While in general this can be viewed qualitatively true, no attempt to optimize the value of the portfolio weights is made. Similar observations are valid for all the models in this class.
However, one should not discard this class of portfolios as being less sophisticated and therefore less performant. Occasionally they may provide significant benefits to an investor.
An important member of this class is the equal weighted portfolio. It is one of the most important benchmarks to assess a portfolio performance.
azapy package provides the following “Naïve” portfolio constructions. The weights are periodically rebalanced according to the targeted policy.
Constant weighed portfolio - it includes the case of equal weighted portfolio,
Inverse volatility portfolio - the weights are proportional to the inverse of asset volatilities,
Inverse variance portfolio - the weights are proportional to the inverse of asset variances,
Inverse drawdown portfolio - the weights are proportional to the inverse of the asset maximum drawdowns experienced during a predefined historical period.
We had included in this class the “Buy and Hold portfolio”. This is an investment strategy where the initial asset positions (number of shares per portfolio component) are preserved over time (no rebalance). The initial portfolio weights are exogenous to this strategy. Similar to the equal weighted portfolio, the “Buy and Hold portfolio” can be viewed as a performance benchmark.