Double Exponential Moving Average (DEMA)

A moving average that keeps up with a steady trend.

What it measures

An EMA trails a trend. An EMA of that EMA trails it twice as far. Taking the second away from twice the first cancels the lag, so on prices rising by the same amount every bar the DEMA sits on the close. Patrick Mulloy published it in 1994. It is not a double smoothing: it is smoother than the price but less smooth than an EMA.

How traders read it

Traders use it where an EMA would turn too late: crossings of price, or of a slower average, come sooner. The price of that speed is more false turns in a sideways market.

Formula

EMA1 = EMA(CLOSE, periods)
EMA2 = EMA(EMA1, periods)
DEMA = 2 × EMA1 − EMA2

Each EMA starts from the mean of its first periods values, as TA-Lib’s DEMA does.

Settings and output

Call: dema(periods)

Setting	Chart default
`periods`: Periods	20

What you get: a number.

Warm-up: with dema(20), it asks for 0 bars of history and a run-up of 120, and its first value came on bar 39 of the example data.

Chart

Double Exponential Moving Average drawn by its own code on example prices

The prices are made up, hourly, and the same on every page, so the chart shows how the indicator behaves, not how a market behaved.

In an algorithm

function onIntervalClose() {
  const fast = dema(20);
  if (fast !== null && CLOSE > fast && OPEN_POSITIONS_COUNT === 0) {
    openPosition("BUY", { quantity: 10000, stopLossPips: 30 });
  }
}