Skip to main content
Track Linear regression
41 / 43
Search the archiveLibrary
Library

2017issue C1147

Constructing Calmar ratio and linear regression baselines

Assemble a monthly Calmar ratio over a 36-month window and a least-squares linear forecast from the same ordered market series. Editorial guidance is to hold both specifications fixed when an in-sample baseline is compared with a later out-of-sample result.

  • The Calmar ratio is constructed on a monthly sampling interval by dividing average return over the most recent 36 months by maximum drawdown over that same window.
  • A negative constructed Calmar ratio indicates that the evaluated system or trader recorded a negative result over the last three years.
  • Linear regression relates an independent variable to a dependent variable so the fitted equation can forecast later values, with the least-squares method placing the curve to minimize the sum of squared deviations.
  • Editorial guidance is to keep both specifications unchanged when the in-sample baseline is compared with a later out-of-sample result.
Entries in this reading2 entries
Calmar ratio1 readingCurrent trackLinear regression112 readings

Two constructions from one series

Editorial framing: assemble two complementary quantitative constructions from the same ordered market series, then hold those specifications fixed when an in-sample baseline is compared with a later out-of-sample result.

The first construction is the Calmar ratio, a monthly risk-adjusted score formed by dividing average return over the last 36 months by maximum drawdown over that same window. The second is linear regression, a fitted mapping from an independent series onto a dependent series that yields a forecast of later values.

SPY 217-222 call debit spread: quote-type returns on May 1

Three quote-type rows from the source table show the same $230 entry debit and $270 max profit, with realized profit rising from $242 (Natural) to $256 (Mid Quote) to $271 (Optimistic). A trader should read that as a 105.2% to 117.8% return on the same $230 risk, not as a change in the structure itself. Numbers are taken from the Figure 2 quote table, not redrawn from the page layout.
Three quote-type rows from the source table show the same $230 entry debit and $270 max profit, with realized profit rising from $242 (Natural) to $256 (Mid Quote) to $271 (Optimistic). A trader should read that as a 105.2% to 117.8% return on the same $230 risk, not as a change in the structure itself. Numbers are taken from the Figure 2 quote table, not redrawn from the page layout.SPY 17Jun16 217-222 call debit spread · as of May 1, 2017

Entry debit, max profit and max risk are identical across quote types; only the marked profit and the greeks change. Downside and upside breakevens are both 219.30.

Constructing the Calmar ratio

The Calmar ratio is constructed by dividing the average rate of return over the most recent 36 months by the maximum drawdown recorded over that same period. Construction of the Calmar ratio is usually carried out on a monthly sampling interval.

Maximum drawdown is the drawdown figure that appears in the Calmar denominator over the same 36-month window used for average return. Where a mean is needed, it is the sum of observed values divided by the number of observations in the sample.

Constructing the linear forecast

Linear regression is constructed by relating an independent variable to a dependent variable so the fitted equation can be used as a forecast of later values.

A least-squares construction places the fitted curve so that the sum of squared deviations from the observed points is minimized.

After a regression is fitted, R-squared reports the share of variation in the dependent series accounted for by the fitted equation and serves as a relative measure of fit.

Pearson's correlation measures linear association between two series and takes values between +1 and -1, from total negative association to total positive association.

Keeping the specifications fixed

Editorial guidance: once the Calmar window, the monthly sampling interval, the independent and dependent series, and the least-squares fitting rule are chosen, keep those specifications unchanged when the in-sample baseline is compared with a later out-of-sample result.

Editorial reading: the Calmar ratio and the linear forecast are complementary views of the same ordered series. One is a 36-month risk-adjusted score. The other is a fitted forecast whose relative fit is described by R-squared and whose linear association can be described by Pearson's correlation.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
41 of 43 in the Linear regression track
201962-68 pp.Next on Linear regressionPair-trade layer construction versus average-spread managementPair construction can add discrete layers at successive spread prices rather than treat the blended average-spread-price as the working position.
All readings on this track · 43 readings
  1. 1990Constructing dollar baselines from rates, inflation, and residuals
  2. 1990Constructing a nominal index value from forward earnings and fitted yield
  3. 1990Constructing a nominal index price from earnings and a fitted yield
  4. 1990Endpoint-pinned price paths are not forecasts
  5. 1990Constructing least-squares polynomial smoothers
  6. 1991Endpoint growth rates versus linear-regression consistency
  7. 1991Out-of-sample checks for linear growth fits
  8. 1991Trend as persistence, not a straight line
  9. 1991Quadratic trend, residual oscillator, and a secondary cycle calendar
  10. 1991Time-origin offset and residual-price divergence on a quadratic least-squares fit
  11. 1991A least-squares trendline from ordered prices
  12. 1992Constructing log-linear growth and reliability screens
  13. 1992Constructing log-linear growth-rate baselines
  14. 1992Next-session high, low, and close from rolling linear regression
  15. 1993Auditing an index price-earnings multiple with short-rate regression
  16. 1994Regression-seeded nested exponential price filter
  17. 1994Constructing the double exponential average from lag cancellation
  18. 1994Evaluating money supply as a linear leading-index baseline
  19. 1995Constructing least-squares trend channels
  20. 1995Linear baseline holdout checks for annual bill-rate forecasts
  21. 1995Projection bands from high and low regression slopes
  22. 1995Evaluating a least-squares end-point moving average on a known test series
  23. 1996Constructing an endpoint moving average from a least-squares line
  24. 1996Scoring equity path consistency with a k-ratio overlay
  25. 1996Evaluating month-end yield gaps for equity regimes
  26. 1996Constructing session-indexed standard error bands
  27. 1998Evaluating linear regression baselines for index valuation
  28. 1998R-squared as a two-state trend filter from a price-time fit
  29. 2000Second-order moving-average lag correction
  30. 2002Price regression line versus beta for index tracking
  31. 2003Regression slope with an r-squared trend confidence gate
  32. 2003Constructing finite-volume-element divergence with slope comparison
  33. 2004Building a daily score from regression, retracement, and volume
  34. 2004Constructing least-squares trendlines from ordered prices
  35. 2007Rectangle breakout targets beyond height
  36. 2007Confirming a price trend with regression slope and r-squared
  37. 2008A linear-regression angle assembled as one trend filter
  38. 2010A two-state swing machine from four running extremes
  39. 2016Score oil-complex tightness before divergence or regression
  40. 2017Nikkei-yen intermarket divergence as a regime case study
  41. 2017Constructing Calmar ratio and linear regression baselines
  42. 2019Pair-trade layer construction versus average-spread management
  43. 2020A convolution slope built from nested linear regression
By fieldStatistical tests59 readingsBy year2017100 readings
All 112 readings tagged Linear regression
Also on Linear regression5 readings
1985Beta and alpha as chosen regression parameters1986Constructing a commodity channel index and a regression price channel1988Constructing breadth and average trend states1988Least-squares construction of channel trendlines1989A precious-metal price as a changing intermarket equation