Skip to main content
Track Volatility forecast
2 / 6
Search the archiveLibrary
Library

1995issue C111-5

A short-to-long historical-volatility ratio as a regime-gate

Build historical volatility twice with one locked annualization-day-count: a 10-observation short-window and a 100-observation long-window. Their volatility-ratio is a regime-gate at the 0.5 compression-threshold. Editorial: that label is a place a later expansion forecast can occupy.

  • A 10-observation short-window and a 100-observation long-window use the same log-return standard-deviation recipe and share one annualized percent scale.
  • The volatility-ratio is the short-window series divided by the long-window series. A reading at or below 0.5 is the compression-threshold.
  • An example regime-gate allows a close-versus-20-session exponential-average entry only if that ratio was at or below 0.5 within the prior 10 sessions.
  • If the long-window is treated as roughly stable, a short-window well below that baseline is presented as a cue to anticipate higher subsequent volatility.
Entries in this reading2 entries
Current trackVolatility forecast6 readingsHistorical volatility analysis47 readings

Two windows on one percent scale

A 10-observation historical-volatility reading is formed from the standard deviation of log close-to-close ratios, then scaled by the square root of 365 and by 100. Each log-return is the natural logarithm of the current close divided by the prior close.

The same log-return standard-deviation construction with a 100-observation lookback produces a slower historical-volatility series on the same annualized percent scale. The illustrated short-window and long-window lookbacks are 10 and 100 sessions.

The volatility-ratio and the compression-threshold

Dividing the 10-observation series by the 100-observation series yields a unitless volatility-ratio. A value below 0.5 means the short-window is less than half the long-window.

That 0.5 level is the compression-threshold. A short-window reading below 50 percent of the long-window reading is the stated cue to anticipate unusually high subsequent volatility.

Yen 10-day/100-day historical-volatility ratio

The short-to-long ratio drops through one-half in mid-June 1995 and again briefly in mid-July, which is the compression regime the tip uses as a gate. Values were read off the MetaStock Japanese-yen pane dated 8 September 1995; the source printed no table.
The short-to-long ratio drops through one-half in mid-June 1995 and again briefly in mid-July, which is the compression regime the tip uses as a gate. Values were read off the MetaStock Japanese-yen pane dated 8 September 1995; the source printed no table.Japanese yen · daily · 1995-05-01T00:00:00.000Z to 1995-09-08T00:00:00.000Z

Both windows use Std(log close-to-close)×√365×100, so those factors cancel in the plotted ratio. The 0.5 line is the gate stated in the tip, not a fitted level. Screenshot readings are only good to about 0.05.

One annualization-day-count for both windows

The same construction is shown with two annualization-day-count conventions: a 365-day calendar count and a 260-day trading-day count. Editorial: lock one convention so the short-window and the long-window stay on one scale before the ratio is formed.

One worksheet form of the same two-window construction multiplies each window's root-mean-square log move by the constant 7.22 instead of computing the square root of 365 divided by 7.

A parameterized function can take lookback length and annualization-day-count as inputs, scale log-return standard deviation by the square root of that day-count divided by 1 on daily bars or 7 on weekly bars, and express the result in percent.

A regime-gate for a separate trend rule

An example entry rule requires a close crossing a 20-session exponential average and, within the prior 10 sessions, a volatility-ratio at or below 0.5.

Editorial: that volatility-ratio condition is a regime-gate. A separate trend rule consults it before an entry is allowed. The ratio labels a regime. It does not replace the trend rule.

The stated follow-on for a compressed short-window

If long-window volatility is treated as roughly stable, a short-window falling well below that baseline is presented as a condition that should be followed by higher volatility. Otherwise the long-window itself would fall.

Editorial: treat a volatility-ratio below the compression-threshold as a regime label that a later expansion forecast can occupy.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
2 of 6 in the Volatility forecast track
20011-5 pp.Next on Volatility forecastConstructing a variable-interval average from a difference-oscillator or volatility forecastA variable-interval average computes, on each bar, a recommended simple-average length from close-to-close difference intervals, a history window used for high-low scaling, and a pair of shortest and longest allowed lengths.
All readings on this track · 6 readings
  1. 1994Constructing hourly index futures lattices from live volatility
  2. 1995A short-to-long historical-volatility ratio as a regime-gate
  3. 2001Constructing a variable-interval average from a difference-oscillator or volatility forecast
  4. 2006Percent-scale average true range for comparable range
  5. 2007Historical compression and implied slope as a futures regime map
  6. 2013GARCH and a volatility rank as market-regime classifiers
By fieldOptions and volatility171 readingsBy year199575 readings
All 6 readings tagged Volatility forecast
Also on Volatility forecast5 readings
2006Percent-scale average true range for comparable range2007Historical compression and implied slope as a futures regime map1994Constructing hourly index futures lattices from live volatility2001Constructing a variable-interval average from a difference-oscillator or volatility forecast2013GARCH and a volatility rank as market-regime classifiers