Software

Bee, M. and Schiavo, S. (2017) "Powerless: gains from trade when firm productivity is not Pareto distributed", Review of World Economics, doi: 10.1007/s10290-017-0295-z

The main Matlab functions contained in the .zip archive compute welfare gains associated with a trade liberalization under a Pareto, lognormal and Weibull productivity distribution.

  • Routines used in the paper "Pareto versus lognormal: a maximum entropy test". The Matlab functions have been written by Marco Bee (University of Trento). Please acknowledge use of the routines by quoting the article as:

Bee, M. and Riccaboni, M. and Schiavo, S. (2011) "Pareto versus lognormal: a maximum entropy test", Physical Review E, 84, 026104.

The .zip archive contains the following Matlab functions:

    • MEd: performs ME estimation for a population y

    • ME: performs ME estimation for the tail of the population y, and computes a llr-test for discriminating between Pareto and lognormal, as well as the UMPU test [updated December 2012]

    • MEplot: plots the p-value of the ME and UMPU tests [updated December 2012]

    • maxent_uni: fits ME using power basis functions (courtesy of Ximing Wu from Texas A&M University; a number of other useful functions are available on his website)

    • maxent_uni_g: fits ME using constraints on logarithmic moments (a slight modification of the original function written by Ximing Wu)

Vaona, A. and Schiavo, S. (2007) "Nonparametric and semiparametric evidence on the long-run effects of inflation on growth", Economics Letters, 94(3), 452-458

  • Here you can download another .zip archive containing a series of Matlab functions I have written some time ago. The archive contains the following functions:

    • lrratio: performs likelihood ratio tests of VAR models to determine the optimal lag length (allows non-contiguous lags)

    • aicsbc: computes AIC & SBC for a VAR model to help selecting the model specification

    • lrratio_exog: performs likelihood ratio tests of VAR models to test exclusion restrictions among exogenous variables

    • mv_qstat: computes the multivariate Ljung-Box Q test for white noise residuals after a VAR model

    • mv_sacf: finds multivariate sample autocorrelation & cross-correlation coefficients and plot them

    • omnibus: computes the Omnibus Test for Univariate & Multivariate Normality (Doornik & Hansen, 1994)

    • rollcorr: computes and plots rolling correlation coefficients

    • rollvar: computes and plots rolling variances

    • vech_up: creates a column vector by stacking the columns of a matrix

    • vech_upc: creates a column vector by stacking the (column) elements of a matrix above the main diagonal

    • vech_upr: creates a column vector by stacking the (row) elements of a matrix above the main diagonal