spseudor2 --- Calculates goodness-of-fit measures in spatial autoregressive models
spseudor2 calculates two so-called pseudo R2 measures to assess goodness of fit in spatial autoregressive models estimated by spatial two stage least squares or spatial GMM. The first measure is computed as the square of the correlation between the predicted and observed values of the dependent variable. The other one is calculated as the ratio of the variance of the predicted values to the variance of the observed values of the dependent variable.
In calculating the predicted values of the dependent variable, spseudor2 takes into account the fact that the spatially lagged dependent variable is endogenous.
N.B.: spseudor2 is intended to work after ivregress and ivreg29 and requires at least Stata 10.1.
Also, spseudor2 assumes that the spatially lagged dependent variable precedes all other endogenous variables, if any, and is the first variable of the listed right-hanside variables in the model.
+---------+ ----+ Options +----------------------------------------------------------
wmat(name) indicates the name of the spatial weights matrix used in the spatial autoregressive model estimation. The matrix must have been saved to a Mata file.
spseudor2 saves the following results in r():
Scalars r(sqcorr) The square of the correlation between the predicted and obser > ved values of the dependent variable r(varRatio) The ratio of the variance of the predicted value to the varia > nce of the observed value of the dependent variable
. spseudor2, wmat(winvecon)
Go to http://stasacode.com for more examples as to how to add the calculated goodness-of-fit measures to estimation results for model comparison purposes.
P. Wilner Jeanty, Dept. of Agricultural, Environmental, and Development Economics, The Ohio State University
Email to email@example.com