Volatility Target Allocator

Volatility Target Allocator holds a long position in an index CFD and decides
how large that position should be from a forecast of volatility, not of
price. When the market is calm it holds more; when it turns violent it holds
less; below the 200-day average it holds nothing.

It is a risk-sizing expert, not a signal generator. It never sells short and
it does not try to call the next move.

WHAT IT DOES

Once a week, after a daily bar closes:

1. Forecasts next week's volatility with a HAR regression on yesterday's,
   last week's and last month's high-low range volatility, refitted from
   scratch on up to six years of daily bars at every resize.
2. Sizes the position as Target Volatility divided by that forecast, capped.
3. Stands aside while price is below its 200-day average. Between resizes the
   average is still checked every day, but only to leave: a break is exited
   the same day, and re-entry waits for the next resize.
4. Adjusts the position, ignoring changes too small to be worth the spread.

WHAT THE RESEARCH SHOWED

NAS100 index CFD, February 2016 to October 2026, daily bars, walk-forward,
with the spread and the broker's quarterly roll adjustment charged:

                            annual   max drawdown   return / drawdown
  Buy and hold, full size    16.6%      45.1%            0.37
  This EA, 15% target        12.8%      18.9%            0.68

It does not add return. It earns about four points a year less than holding
the index outright, and in exchange its worst drawdown is well under half.
That is not just a smaller position: holding the same average exposure fixed
keeps buy-and-hold's 0.37 ratio. The EA gets to 0.67 by being small at the
right times rather than small all the time.

TARGET VOLATILITY: 15% UP TO 40%

The target is chosen from a list - 15, 20, 25, 30, 35 or 40% - and every
setting up to 40% may be used. Raising it raises return and drawdown
together; the ratio between them stays at 0.67-0.68 on every setting,
because the maximum weight and the equity brake move with the target
automatically.

  target   annual   max drawdown   max weight   brake
   15%     12.8%       18.9%           2.0        35%
   20%     16.8%       25.2%           2.0        40%
   25%     20.6%       30.5%           2.0        50%
   30%     25.2%       37.8%           3.0        60%
   35%     29.2%       43.0%           3.0        70%
   40%     32.3%       48.5%           3.0        80%

At 35% and 40% the drawdown is as deep as holding the index itself, or
deeper. Choose them only if you could sit through that, and allow for live
drawdowns running deeper than any backtest. The default, 15%, is the
setting for most accounts.

SAFETY

- Equity brake: flattens on a severe drawdown from the peak, then waits for
  a fresh trend signal before trading again
- Margin cap: a resize never takes the margin in use past 50% of equity
- Spread guard and session check: a resize waits rather than paying a
  rollover spread or sending orders into a closed market; refused orders
  are retried at most once per chart bar
- One copy per symbol: a second chart with the same magic number refuses to
  start, so the position is never bought twice
- State survives a terminal restart
- Every reason for not trading is printed to the journal, so a flat stretch
  can always be explained

WHERE IT WORKS

Recommended: equity index CFDs with a long-term upward drift that clearly
exceeds the cost of holding them - NAS100 in the research. It can be
attached to any symbol, but on an instrument with no upward drift, such as a
currency pair, there is nothing for it to size and it should not be expected
to make money.

HOW TO USE

1. Attach it to an index CFD chart. H1 is recommended; it reads daily bars
   internally whatever the chart shows.
2. It needs 250 daily bars before the first trade, so let the history
   download first.
3. Keep Exposure At Weight 1.0 at 100% on an account dedicated to it, and
   lower it if the account is shared with other strategies.
4. Run it on a demo account for a few weekly resizes before going live.

NOTES

- The figures above come from a research backtest on daily data
  (2016-2026), checked against the MT5 Strategy Tester (2019-2026). The
  tester charges no rollover adjustment on futures-based index CFDs, so its
  results run 2-4 points a year higher than a real account would see.
- The account must be large enough for the position to be expressed. On an
  index CFD traded in 0.1-lot steps, a 10,000 account can only adjust in
  steps of about a third of its equity.
- This is a position-sizing tool, not a promise of profit. Past results do
  not guarantee future results.
Другие продукты этого автора
Pivot Box Range draws daily pivot levels and a high-low range box for each day, and it fixes the one thing that quietly breaks most pivot indicators: the broker's short days. THE SHORT DAY PROBLEM Most brokers open a Sunday candle that lasts only an hour or two, and some holidays produce half sessions. A pivot indicator that treats those as a full day calculates Monday's levels from almost no data, and the levels land in the wrong place for the whole week. This indicator merges a short day i
FREE
Session Box Range draws the Asian, European and US trading sessions as high-low boxes, so you can see at a glance where each session opened, how far it travelled and where it closed. The range of the most recent session is printed on the on-chart toggle buttons, so the number stays readable no matter how far you zoom or scroll. FEATURES - Session hours are entered in GMT, and the broker's server offset is detected   automatically, so the same settings work on any broker and survive the   ser
FREE
Currency Strength Ribbons shows the eight major currencies - USD, EUR, GBP, JPY, CHF, CAD, AUD and NZD - as smooth lines in one window, each measured against all the others at once from the 28 pairs that link them. HOW IT DIFFERS Most strength meters add up pair changes one pair at a time. That counts every move twice: when EURUSD rises, the euro gets full credit and the dollar full blame, whatever the other 26 pairs say about which of the two actually moved. Currency Strength Ribbons solves
Фильтр:
Нет отзывов
Ответ на отзыв