Day-Ahead Crude Oil Price Forecasting Using a Novel Morphological Component Analysis Based Model

Algorithm 1

A MCA based hybrid methodology.

𝑁← the initial decomposition scale

𝑟_𝑡=∑^𝑁_𝑖𝑟_𝑖,𝑡  {Component extraction}

while  𝑖<𝑁  do

 𝑠_𝑖← Statistical tests of the null hypothesis that the DGP is i.i.d or linear {Feature identification}

 if  𝑠_𝑖≥0  then

  𝑟_𝑡=𝛿_𝑡+∑^𝑚_𝑖=1𝜙_𝑖𝑟_{(𝑡−𝑖)}+∑^𝑛_𝑗=1𝐿𝜃_𝑗𝜀_{(𝑡−𝑗)}+𝑎_𝑡  {Linear data modeling}

 else

  𝑟_𝑡=∑^𝑝_𝑖=1𝜔_𝑖𝑟_{(𝑡−𝑖)}+𝑏_𝑡  {Nonlinear data modeling}

 end if

end while

𝑟(𝑡)=∑^𝑁_𝑖𝑟_𝑖,𝑡  {Forecast reconstruction}