Statgraphics for Windows
Analysis summary report for Series #2
Random walk model with drift 
Analysis Summary

Data variable: Y2

Number of observations = 101
Start index = 1.0
Sampling interval = 1.0

Forecast Summary
----------------
Nonseasonal differencing of order: 1

Forecast model selected: ARIMA(0,1,0) with constant
Number of forecasts generated: 12
Number of periods withheld for validation: 20

            Estimation      Validation
Statistic   Period          Period
--------------------------------------------
MSE         0.886679        1.02186         
MAE         0.762752        0.835759        
MAPE        3.85985         2.263           
ME          -0.00515478     0.381454        
MPE         -0.0865215      1.00468         

                            ARIMA Model Summary
Parameter           Estimate        Stnd. Error     t               P-value
----------------------------------------------------------------------------
Mean                0.259231        0.105278        2.46234         0.015531
Constant            0.259231        
----------------------------------------------------------------------------
Backforecasting: yes
Estimated white noise variance = 0.886679 with 99 degrees of freedom
Estimated white noise standard deviation = 0.941636
Number of iterations: 2
In this model (random walk with drift), the constant term (0.259231) represents the estimated trend in the series--i.e., the average period-to-period change.