GetNewLambdaWeight: Portfolio Weights of the Lambda-strategy

In RelValAnalysis: Relative Value Analysis

Description

The function computes the portfolio weights of the lambda strategy.

Usage

GetNewLambdaWeight(pi.current, mu.next, energy, lambda = 0.5)

Arguments

pi.current	current portfolio weight.
mu.next	market weights for the next period.
energy	a non-negative number. The free energy of the previous period.
lambda	a number between 0 and 1. The default is 0.5.

Details

The lambda-strategy is a simple energy-entropy strategy that depends on a parameter lambda between 0 and 1, see Section 6.1 of Pal and Wong (2013) for more details. Given the portfolio weights of the previous time period, the portfolio weights of the next period is a convex combination of the previous weights and the current market weights, chosen in such a way that the drift term is a constant proportion of the free energy.

In this implementation an SDE approximation is used. Hence the function is accurate only when lambda is small (say <= 0.5).

Value

A portfolio weight vector.

References

Pal, S. and T.-K. L. Wong (2013). Energy, entropy, and arbitrage. arXiv preprint arXiv:1308.5376.

See Also

GetLambdaWeight

Examples

pi.previous <- c(1/3, 1/3, 1/3)
mu.next <- c(0.5, 0.3, 0.2)
energy <- 0.05
lambda <- 0.3

GetNewLambdaWeight(pi.previous, mu.next, energy, lambda)

RelValAnalysis documentation built on May 2, 2019, 3:09 a.m.