ARIMA structure

[tomasza]

I am trying to find the arima structure of my dataset. After testing for and finding a unit root, i use first differences. However now I am trying to find a way to find an appropriate arima structure. My corrolelogram shows no clear pattern. Only small bars for the 3rd, 14th and 16th auto and partial autocorrelation. What does this mean for my ARIMA structure. Do I really have to try all structures by trail and error and compare schwarz or akaike values? testing all possible structures combining ar and ma up to 5 lags, my best fit is an MA(1) structure. However this MA(1) term is highly insignificant.