atra.mrio package¶
Submodules¶
atra.mrio.create_mrio module¶
atra.mrio.ras_method module¶
RAS Method
Purpose¶
Estimate a new matrix X with exogenously given row and column totals that is a close as possible to a given original matrix X0 using the Generalized RAS (GRAS) approach
Usage¶
X = gras(X0, u, v) OR [X, r, s] = gras(X0, u, v) with or without eps included as the fourth argument, where
Input¶
- X0 = benchmark (base) matrix, not necessarily square
- u = column vector of (new) row totals
- v = column vector of (new) column totals
- eps = convergence tolerance level; if empty, the default threshold is 0.1e-5 (=0.000001)
Output¶
- X = estimated/adjusted/updated matrix
- r = substitution effects (row multipliers)
- s = fabrication effects (column multipliers)
References
- Junius T. and J. Oosterhaven (2003), The solution of updating or regionalizing a matrix with both positive and negative entries, Economic Systems Research, 15, pp. 87-96.
- Lenzen M., R. Wood and B. Gallego (2007), Some comments on the GRAS method, Economic Systems Research, 19, pp. 461-465.
- Temurshoev, U., R.E. Miller and M.C. Bouwmeester (2013), A note on the GRAS method, Economic Systems Research, 25, pp. 361-367.