AR(2) One-Step Forecast

Problem

Compute the one-step-ahead forecast of a fitted AR(2) model: c + phi1 * y_t + phi2 * y_{t-1}, where y_t is the most recent observation and y_prev the one before it.

All coefficients are scalars and the forecast is returned as a float.

Hint

An AR(2) forecast is a linear combination of the two most recent values plus the intercept.

Test cases

case 1

in: [0.5,0.5,0.25,2,1]

expected: 1.75

case 2

in: [0,0.7,-0.2,5,4]

expected: 2.7

case 3

in: [1,0,0,3,9]

expected: 1

case 4

in: [0.2,0.5,0.3,0,0]

expected: 0.2

Starter codePython

def ar2_one_step_forecast(c, phi1, phi2, y_t, y_prev):
    # Your code here
    pass

4 test cases · run your code against them in the editor