The RATS Software Forum

FRB Cleveland's working paper website just post this paper by Todd Clark and Francesco Ravazzolo "The Macroeconomic Forecasting Performance of Autoregressive Models with Alternative Specifications of Time-Varying Volatility" It compares AR models with various time-varying volatility specif...

As for Stock and Watson's research memorandum "Distribution of quarterly values of GDP/GDI across months within the quarter
Background", is there a replication here?

http://www.princeton.edu/~mwatson/mgdp_gdi.html

I see. I thought (L_t = exp(logI_t) + .....exp(logI_t-q) or L=I_t+...I_t-q; can be combined with any one of (7.1, 7.2, 7.2), which are I=H+Z; I=Hz; LogI=H+z As you pointed out, DISAGGREGATE 's option LOGLIN actually indicates a combination of L_t = exp(logI_t) + .....exp(logI_t-q) and LogI=H+z; whil...

How to calculate the roots of a specific polynomial?
I mean this poly is not from a regression but I need to input the coefficients manually.

Thanks

About example history.prg, how the sequence is chosen? Is there a reference for that? * HISTORY.PRG * * HISTORY.PRG * Manual Example 10.5 * ********************************************************************* * * Set Up Data * open data oecdsample.rat calendar(q) 1981 data(format=rats) 1981:1 2006:...

real exchange rate should be kept in log levels or log difference?

How g is chosen?

What is the parameter setting to correspond to DISTRIBUTE procedure where no related series are used? MODEL=LINEAR,MAINTAIN=SUM plus the choice for the time series model. What to put under "#" Another question is about option LOGLINEAR. My understanding is that this "loglinear" ...

There is another example using denton method to distribute data with a related series. Do we need to multiply the distrib by 4? set distrib =distrib*4 cal(q) 1998 all 2000:4 data(unit=input) / quarter annual 98.2 100.8 102.2 100.8 99.0 101.6 102.7 101.5 100.5 103.0 103.5 101.5 . . . 4000 . . . 4161....

Also in UG, there is an example using DISTRIB cal(m) 1947 open data haverexample.rat data(for=rats, verbose) 1947:1 2006:4 gdp @distrib(factor=3) gdp gdpm set gdpm = 3*gdpm It says we need to multiply the result by 3 to maintain the original level. In this case, M1+M2++M3 is not equal to the origin...

Also in UG, there is an example using DISTRIB cal(m) 1947 open data haverexample.rat data(for=rats, verbose) 1947:1 2006:4 gdp @distrib(factor=3) gdp gdpm set gdpm = 3*gdpm It says we need to multiply the result by 3 to maintain the original level. In this case, M1+M2++M3 is not equal to the origina...

What is the parameter setting to correspond to DISTRIBUTE procedure where no related series are used?

This data series fits ARIMA(2,1,2) (2,1,3) (2,1,0) well but when fit to ARIMA(2,1,1), BOXJENK(DIFFS=1,CONST,MAXL,AR=2,MA=1) FF(860) it shows Box-Jenkins - Estimation by ML Gauss-Newton NO CONVERGENCE IN 34 ITERATIONS LAST CRITERION WAS NA SUBITERATIONS LIMIT EXCEEDED. ESTIMATION POSSIBLY HAS STALLED...

What I got wrong is the use of (1-%zlag(t,1) to represent (1-L) part after calculating |C(w)|^2, when I have the equation like (1-Phi(L))yt=THETA(L)(1-L)et Now I define the EQUATION using parameters in Phi(L) and Theta(L) by expanding Theta(L)(1-L) equation(noconstant,coeffs=||phi1, phi2, theta1-1,-...

Equation: yt=PSI(L)(et-et-1) and assume we do not have MA terms here and only AR(2) terms. I correct the squaring error in previous post The spectrum are still not the same. What is the reason? Which calculation is wrong? equation(noconstant,coeffs=||@@, @@,-1||) arma21 y 2 1 frequency 1 512 com sca...