svar long run restrictions problem

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

svar long run restrictions problem

Deborah Sy
This post has NOT been accepted by the mailing list yet.
To the Stata Community,

I am trying to run a 9-variable SVAR model using long run constraints in lieu of Blanchard and Quah. Using the formula (n^2-n)/2, I am left with 36 constraints. I was able to make 36 restrictions in the matrix but I cannot seem to generate the SVAR result. The following is my log-file:

svar var1 var2 var3 var4 var5 var6 var7 var8 var9, lreq(lr) lags(1/3)
With the current starting values, the constraints are not sufficient for identification
The constraints placed on C are
1999: [c_1_2]_cons = 0
1998: [c_1_7]_cons = 0
1997: [c_1_9]_cons = 0
1996: [c_2_5]_cons = 0
1995: [c_2_7]_cons = 0
1994: [c_2_8]_cons = 0
1993: [c_3_4]_cons = 0
1992: [c_3_5]_cons = 0
1991: [c_3_6]_cons = 0
1990: [c_3_7]_cons = 0
1989: [c_3_8]_cons = 0
1988: [c_3_9]_cons = 0
1987: [c_4_1]_cons = 0
1986: [c_4_3]_cons = 0
1985: [c_4_5]_cons = 0
1984: [c_4_6]_cons = 0
1983: [c_4_7]_cons = 0
1982: [c_5_2]_cons = 0
1981: [c_5_3]_cons = 0
1980: [c_5_4]_cons = 0
1979: [c_5_8]_cons = 0
1978: [c_5_9]_cons = 0
1977: [c_6_1]_cons = 0
1976: [c_6_3]_cons = 0
1975: [c_6_7]_cons = 0
1974: [c_6_9]_cons = 0
1973: [c_7_2]_cons = 0
1972: [c_7_8]_cons = 0
1971: [c_8_1]_cons = 0
1970: [c_8_2]_cons = 0
1969: [c_8_3]_cons = 0
1968: [c_8_4]_cons = 0
1967: [c_8_7]_cons = 0
1966: [c_9_1]_cons = 0
1965: [c_9_3]_cons = 0
1964: [c_9_6]_cons = 0
These constraints place 36 independent constraints on C
The order condition requires at least 36 constraints
r(498);

I am currently using StataIC 12 for Windows (32-bit).

Does anyone know what went wrong here? Your help is truly appreciated.
Loading...