I have a momentum signal that gives best predictive power at 3 months and a valuation signal that gives best predictive power at one year. If I combine for a 3 month horizon by "interpolating" the valuation signal i.e. dividing the valuation signal by 4, what if any statistical problems will this generate for the portfolio optimization?

  • $\begingroup$ Sharpe Ratio is annualized on e.g. daily data using Brownian motion stddev by multiplying the daily value by sqrt(365). using this approach would mean you should divide the valuation signal not by 4 but by sqrt(4)=2 $\endgroup$ – MarianP Jul 11 '18 at 11:20
  • $\begingroup$ No by signal I meant expected return derived from a signal. The mean is linear in time. $\endgroup$ – Andrew Beaven Jul 23 '18 at 14:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.