Seasonal and directional strength

predictMove (vars Data, int Length, int Horizon, var Percentile): var

Predicts the magnitude of a price movement, based on statistical analysis of price changes in the Data series. Returns the given Percentile of Data changes within the given time Horizon. Example: if the function returns 10 at Horizon = 20 and Percentile = 95, then in 95% of all cases the Data value moved by 10 or less in any direction within 20 bars (and therefore moved by more than 10 only in 5% of all cases). For statistical significance Length should be large compared to Horizon

predictSeason (vars Data, int Length, int Horizon, int Season): var

Predicts the expected price movement within a given time Horizon, based on the current time and seasonal movement in the Data series. Example: if the function returns 1.5 at Horizon = 20 and Season = 4, then the Data series is expected to rise by 1.5 within 20 bars from now, based on the annual movement at the same date in previous years. If Horizon == 0, the function returns no price move, but the average seasonal Data value at the current date. For statistical significance the Data series should cover at least 3 or 4 seasons. 

Returns:

Predicted data change / average seasonal data.

Parameters:

Data A data series, usually from price functions price(), priceClose() etc.
Length The length of the data series, normally the LookBack period.
Horizon Price movement duration in bars.
Percentile Price movement percentile to return, f.i. 5 for the lowest 5% or 95 for the highest 5%.
Season 1 for daily, 2 for weekly, 3 for monthly, or 4 for annual seasons.

Remarks:

Example:

function run()
{
  StartDate = 2010;
  BarPeriod = 1440;
  LookBack = 300;	
 
  vars Prices = series(price());
  if(!is(LOOKBACK)) {
    LifeTime = 1;
    var Season = predictSeason(Prices,LookBack,LifeTime,2);
    if(Season > 0)
      enterLong();
    else if(Season < 0)
      enterShort();
  }
}

See also:

asset, ROC, predict

► latest version online