I am working on a project that uses ECM model to inspect the short-run dynamics of money supply (m(t)) to loans (l(t)) since both variables are I(1). Excluding the error correction term, is it appropriate to aggregate the coefficients for l(t) to obtain its overall short-run effects? i.e. for al(t-2)+bl(t-1)+cl(t), I obtain total short-run coefficient by a+b+c. I am not sure if this is an appropriate way to follow.
Please, I really need technical assistance from experts here.
Regards,
Eli
yeah you can, because the ecm is useful for estimating both short and long term effect of one time series on the another. ect relates to the fact that last period deviation from long run equilibrium, the error influence its short-run dynamics.
If you have an equation with estimated coefficients which are acceptable and reasonable, why not simulate a "shock" of the exogenous variable the effect of which you are looking for? By comparison with some base variant (simplest: holding other variables constant), one gets a pattern of the influence of this shock over time (for all following t's). What lags you consider as short-run is up to you. This effect can be calculated via mathematics, too, but that could be rather complicated. You can try to variate the coefficients within the confidence intervals (sensitivity analysis) to find out how stable the effects are.
This is not proper way to get total short-run effect you shoul use different model to get it like impulse response as results in shot run vary from period 1 to period 2
Guechari Yasmina , Anton Rainer and Salman Haider thank you all, for your valuable responses. I will try and apply each recommendation.
- Badges
- Science topic
- Similar topics
- Correction