math - In R, Coefficients of MA Function Are Wrong? -

i'm sifting through copy of analysis of financial time series 2nd edition ruey tsay, , 1 of sections involves fitting ma model data (data set here). here's fit exact maximum likelihood according text, insignificant parameters removed:

rt = 0.013 + a(t) + 0.181a(t−1) − 0.121a(t−3) + 0.122a(t−9)

σ(a) = 0.0724

however, when try fit r...

> mew = read.table("m-ew.dat") > arima(mew,order = c(0,0,9),fixed = c(na,0,na,rep(0,5),na,na),method = "ml") call: arima(x = mew, order = c(0, 0, 9), fixed = c(na, 0, na, rep(0, 5), na, na),  method = "ml")  coefficients:         ma1  ma2      ma3  ma4  ma5  ma6  ma7  ma8     ma9  intercept       0.180    0  -0.1318    0    0    0    0    0  0.1373     0.0132 s.e.  0.031    0   0.0362    0    0    0    0    0  0.0327     0.0029  sigma^2 estimated 0.005282:  log likelihood = 1039.1,  aic = -2068.21 

as can see, ma1 coefficients same, ma3 , ma9 different, method = "ml", i.e. maximum likelihood. why this?

also, practical standpoint, while ma2 , ma4-ma8 may 0 (their 95% confidence intervals overlap 0), removing them model raises aic, lowers p-value regards ljung-box test on residuals, , lowers log-likelihood value. worth removing these parameters if such things happen?

in arima 1 can read: "the results different s-plus's arima.mle, computes conditional likelihood , not include mean in model. further..."

and tsay uses s-plus...