The RATS Software Forum

How can I do that please? I would like to work with the resulting coefficient and standard error series from rollreg but instructions such as SET do not recognise them (because of the brackets, I suspect).

Thank you.

No. Unless you did something wrong---clearing the memory should have no effect on a re-run. If you're getting very different results from different runs, then you have an absorbing state in the chain where it apparently either picks 1 or picks 0 for some set of entries and can't move away from thos...

I assume you have a fair number of entries where the probability of the regimes is somewhere in the midrange rather than near 0 or 1. No, actually the regimes are very clearly defined with most probabilities being close to 0 or 1. To see what's going on, I ran my program six times and got two sets ...

I have adapted this for my own work and the output is very sensible, especially w.r.t. impulse responses. I'm using the MCMC program with 20,000 draws (4,000 burn-in) and find that the number of observations in each regime differs quite substantially from estimation to estimation. Should I increase ...

The paper "Inflation and Growth: New Evidence From a Dynamic Panel Threshold Analysis" by Kremer, Back and Nautz (Empirical Economics, 2013) updates Hansen's (1999, 2004) work to address endogeneity in panel threshold models. Alexander Bick makes his data and Matlab code available here: ht...

OK, I reconstructed Pi by multiplying the vector of ECTs with the cointegrating vector and generated the 2x3 matrix I was looking for. Figured out that %VECMPI does not include the estimates of the trends (even though they are in the CV).

Is it possible to replicate this CATS instruction
in RATS? Thank you.

Can you please put me out my misery here. In CATS, this @cats(lags=2,dettrend=cidrift) # myvar1 myvar2 gives me a 2x3 matrix Pi. I am trying to replicate this in RATS with the following: @johmle(lags=2,det=rtrend,cv=cv) # myvar1 myvar2 * compute cv=cv/cv(1) * set trend = t set ect = cv(1)*myvar1+cv(...

Thank you for your contribution. What's the significance of the 30 to 70 range used (in SMPL's and in DO loops)? This sets the range of asset prices for which the deltas and other greeks are calculated. The horizontal axis in the relevant graphs (i.e. plots of a greek against asset price) uses this...

Thank you for all the input. The following shows how to plot option greek parameters against asset prices, maturity, etc. in RATS. The examples correspond to those in the chapter on the Greeks from Hull's Options, Futures and Other Derivatives . This sets the input values for the non-dividend-paying...

You're right, I didn't explain well what I need. Before plotting deltas over time, I need to generate them first, e.g. with something like this: do i = 1,100 compute delta(i)=%cdf((log(price/strike)+expire(i)*(rate-cashflow+.5*sigma**2))/(sigma*sqrt(expire(i)))) end do i The problem with the above i...

I'm trying to do the above but can't figure out how. I have had success plotting delta, theta and gamma against the underlying asset price, as in the following program: smpl 30 70 * * calculate BSM price * @BSOPTION(price=49,strike=50,rate=0.05,sigma=0.2,expire=0.3846) bsm dis "BSM price is&quo...

I'm not using their (Balcilar et al's) data and 'my' likelihood is better behaved. I guess it should be OK.

Thank you for the reference. I can see that Fruhwirth-Schnatter uses the SBC to select the number of regimes in chapter 11. I have to read a couple of the references therein to understand more, as I got confused with the "bridge", "importance", and "reciprocal importance&quo...

I don’t know. Is testing with the SBC not subject to the issue of nuisance parameters? It seems surprising that so much work has gone into resolving this, if the SBC were to be the answer.