The Hodrick–Prescott filter (also known as Hodrick–Prescott decomposition) is a mathematical tool used in macroeconomics, especially in real business cycle theory, to remove the cyclical component of a time series from raw data. It is used to obtain a smoothed-curve representation of a time series, one that is more sensitive to long-term than to short-term fluctuations. (source Wikipedia: https://en.wikipedia.org/wiki/Hodrick%E2%80%93Prescott_filter)

Be aware that this indicator is continuously rechecking the complete history to redraw the curve to be perfectly in line with the data serie. So it is repainting the past and should not be used “as is” for live trading.

Compatible only with PRT v11 and onwards.

defparam drawonlastbaronly=true

//PRC_Hodrick-Prescott filter | indicator
//28.09.2020
//Nicolas @ www.prorealcode.com
//Sharing ProRealTime knowledge

// --- settings 
nobs   =1000 //Number of bars to smooth
lambda =1600 //Higher lambda leads to the smoother data
// --- end of settings 

if islastbarupdate and barindex>max(nobs,lambda) then 
 for i=0 to nobs-1
  $output[i]=Close[i] 
 next

 $a[0]=1.0+lambda
 $b[0]=-2.0*lambda
 $c[0]=lambda
 for i=1 to nobs-3 do 
  $a[i]=6.0*lambda+1.0
  $b[i]=-4.0*lambda
  $c[i]=lambda
 next

 $a[1]=5.0*lambda+1
 $a[nobs-1]=1.0+lambda
 $a[nobs-2]=5.0*lambda+1.0
 $b[nobs-2]=-2.0*lambda
 $b[nobs-1]=0.0
 $c[nobs-2]=0.0
 $c[nobs-1]=0.0

 //Forward
 for i=0 to nobs-1 do
  Z=$a[i]-H4*H1-HH5*HH2
  HB=$b[i]
  HH1=H1
  H1=(HB-H4*H2)/Z
  $b[i]=H1
  HC=$c[i]
  HH2=H2
  H2=HC/Z
  $c[i]=H2
  $a[i]=($output[i]-HH3*HH5-H3*H4)/Z
  HH3=H3
  H3=$a[i]
  H4=HB-H5*HH1
  HH5=H5
  H5=HC
 next

 //Backward
 H2=0
 H1=$a[nobs-1]
 $output[nobs-1]=H1
 for i=nobs-2 downto 0 do 
  $output[i]=$a[i]-$b[i]*H1-$c[i]*H2
  H2=H1
  H1=$output[i]
 //drawpoint(barindex[i],$output[i],1) coloured(255,255,0)
 drawsegment(barindex[i],$output[i],barindex[i+1],$output[i+1]) coloured(255,255,0) style(line,3)
 next
endif

return