Gekko User Manual > Gekko statements

ANALYZE calculates statistics on one or more timeseries or timeseries expressions.

For each expression, Gekko prints out mean, standard deviation, and min and max values. In addition, cross-correlations are computed if there are more than one expression, and the cross-correlations is put into the matrix #corr.

Syntax

analyze <period> variables;

period	(Optional). Local period, for instance 2010 2020, 2010q1 2020q4 or %per1 %per2+1.
variables	A list of variables (timeseries expressions)

•If no period is given inside the <...> angle brackets, the global period is used (cf. TIME).

•If a variable without databank indication is not found in the first-position databank, Gekko will look for it in other open databanks if databank search is active (cf. MODE).

Examples

Analyze the growth rate of the three variables x, y, z:

time 2001 2005;
x = 1.3, 1.2, 1.3, 1.4, 1.2;
y = 1.2, 1.4, 1.7, 1.6, 1.5;
z = 1.6, 1.4, 1.2, 1.3, 1.4;
analyze <2002 2005> pch(x), pch(y), pch(z);

//pch(x)
// 4 observations from 2002 to 2005
// Mean = -1.4881
// St.dev. = 11.2991
// Min = -14.2857 in 2005
// Max = 8.3333 in 2003
//pch(y)
// 4 observations from 2002 to 2005
// Mean = 6.4907
// St.dev. = 14.6300
// Min = -6.2500 in 2005
// Max = 21.4286 in 2003
//pch(z)
// 4 observations from 2002 to 2005
// Mean = -2.6900
// St.dev. = 12.3828
// Min = -14.2857 in 2003
// Max = 8.3333 in 2004
//
//Cross correlation based on 3 variables and 4 observations
//
//#corr
// 1 2 3
// 1 1.0000 0.2683 -0.2018
// 2 0.2683 1.0000 -0.9967
// 3 -0.2018 -0.9967 1.0000

Note

The cross-correlations are computed as Pearson product-moment correlation coefficients.

If you square the cross-correlation matrix (prt multiply(#corr, #corr);), these squared values correspond to the R2 values you obtain by pairwise linear regression between the variables, for instance ols x2 = x1;.

Related statements

OLS