Definition of MA Process / Moving Average Process: MA stands for "moving average." Describes a stochastic process (here, e_t) that can be described by a weighted sum of a white noise error and the white noise error from previous periods. An MA(1) process is a first-order one, meaning that only the immediately previous value has a direct effect on the current value:

e_t = u_t + pu_t-1

where p is a constant (more often denoted q) that has absolute value less than one, and u_t is drawn from a distribution with mean zero and finite variance, often a normal distribution.

An MA(2) would have the form:

e_t = u_t + p₁u_t-1 + p₂u_t-2

and so on. In theory a process might be represented by an MA(infinity). (Econterms)

Terms related to MA Process / Moving Average Process:

• Stochastic Process
• White Noise