svm - time series forecasting using Support Vector Machine -

what attributes used in time series forecasted using svm? have 2 values date , value @ date class know can use -1 , 1 when price gets or down still don't know how plot time series calculate hyperplane

there papers show ways it:

financial time series forecasting using support vector machine

using support vector machines in financial time series forecasting

financial forecasting using support vector machines

i recommend go through existent literature, fun describe easy way (probably not best) it.

let's have n pairs particular date/time of pair , corresponding value. pairs sorted x component.

let's want predict if given , corresponding unknown value go or down (notice use regression , instead try predict value itself).

then train model training set this:

input                         value ======================        ================  y_t0, y_t1, ..., y_ti-1         1 :if y_ti   > y_ti-1, -1 otherwise y_t1, y_t2, ..., y_ti           1 :if y_ti+1 > y_ti,   -1 otherwise y_t2, y_t3, ..., y_ti+1         1 :if y_ti+2 > y_ti+1, -1 otherwise y_t3, y_t4, ..., y_ti+2         1 :if y_ti+3 > y_ti+2, -1 otherwise y_t4, y_t5, ..., y_ti+3         1 :if y_ti+4 > y_ti+3, -1 otherwise 

basically, training algorithm make educated guess of next "tick" in future given glimpse of past. once model trained make prediction feed model n values (where n amount of values used input in training phase) previous value want predict.