runs=1        */ Number of runs
output data   */ Open a file AR.data to output the x(t) generated
prob threshold=0.01       */ Threshold for AR t p-values if AR fit is used
forecast=no               */ Forecast=no, fitted model, random walk
#  */ end of first parameter read loop
sample size=40            */ Sample size
noise model=gaussian      */ gaussian, expon, double tailed expon, unif, log normal
AR lag=2                  */ p for AR(p) or q for MA model
root magnitude=0.7        */ Maximum magnitude of AR(p) roots
frequency=0.1             */ 1st root frequency 0<fr<0.5
innovations std dev=0.0   */ standard devation of e(t)
trend type=stochastic     */ Trend=none, curve, stochastic
trend intercept=1.0       */ Intercept of trend line
linear trend slope=4.0    */ Slope of trend line
slope of quadratic=0.5    */ Slope of quadratic component of curvalinear trend
drift=0.20                */ Drift constant
model fit=AR(4)           */ Model fit=AR(!np) to the data or residuals
detrend=first difference  */ Detrend=no, curve, (first or second) difference
#  */ end of parameter reads for runs>=1
sample size=5000          */ Sample size
noise model=log normal    */ gaussian, expon, double tailed expon, unif, log normal
AR lag=4      */ p for AR(p) or q for MA model
root magnitude=0.7        */ Maximum magnitude of AR(p) roots
frequency=0.4             */ 1st root frequency 0<fr<0.5
innovations std dev=7.0   */ standard devation of e(t)
trend type=stochastic     */ Trend=none, curve, stochastic
trend intercept=1.0       */ Intercept of trend line
linear trend slope=4.0    */ Slope of trend line
slope of quadratic=0.5    */ Slope of quadratic component of curvalinear trend
drift=0.20                */ Drift constant
model fit=AR(7)           */ Model fit=AR(!np) to the data or residuals
detrend=curve             */ Detrend=no, curve, (first or second) difference
#  */ end of parameter reads for runs>=1