GitLab

If ts is a dseries object, then

ts{'Variable_1','Variable_2'} = [];

or

ts{'Variable_@1,2@} = [];

will remove Variable_1 and Variable_2 from ts (if they exist).

If ts is an empty dseries object with a defined range of dates:

ts = dseries(dates('1990Q1'):dates('1990Q4'));

Then the following syntaxes are valid:

1. ts(:) = 1;
2. ts(:) = [1, 2];
3. ts(:) = randn(4,1);
4. ts(:) = dseries(1);
5. ts(:) = dseries([1, 2]);
6. ts(:) = dseries(randn(4,2));
7. ts(:) = dseries(randn(4,2),dates('1950M1'):dates('1950M4'));

Remarks.

[1] In cases 1., 2., 4. and 5. the single observation is replicated to match the number of dates in ts.

[2] In cases 4. to 7. the dates in the right member of the assignment are lost (ie ts.dates is not affected by the dates defined in the right members).

(cherry picked from commit 609f9488)

(cherry picked from commit 667e3085)

Added the possibility to overwrite the time and init member of a dynSeries object (not allowed for freq, nobs and vobs).

(cherry picked from commit b0d6e2b7)

Moved the commented first line, used to declare that the routine has unitary tests, after the declaration of the function, otherwise help is not displayed.

*Examples*

 * ts.A(qq(1950,1):qq(1950,4)) = 1 will put ones for variable A in periods 1950Q1 to 1950Q4

 * ts{'A','B'}(qq(1950,1):qq(1950,4)) = 1 will put ones for variables A and B in periods 1950Q1 to 1950Q4

 * ts{'A','B'}(qq(1950,1):qq(1950,4)) = ones(4,1) will put ones(4,1) for variables A and B and sub-periods 1950Q1-1950Q4

 * ts{'A','B'}(qq(1950,1):qq(1950,4)) = ones(1,2) will put ones(1,2) for variables A and B and each period between 1950Q1 and 1950Q4

Fixed bug in @dynSeries/subsasgn that pops up when length(S) is equal to two and S(1).type is a dot.

Allows Syntax like:

ts.A1(rg) = B

where ts is a dynSeries object, A1 is a variable in the database, rg is a dynDates object and B an array of doubles.

The following syntaxes are now allowed:

ts{'VARIABLE_NAME'}(range) = data;

where range is a dynDates object and data is an array of doubles, or

ts{'VARIABLE_NAME'}(range) = TS{'OTHER_VARIABLE_NAME'}(range)

where ts and TS are dynSeries objects.

As in dynSeries/extract we consider a tensor product.

 + Added the possibility to rename the assigned variables.
 + Added unitary tests.

Following syntax:

ts{'GDP_US'} = ts.{'GDP_US'}.log

applies the log function to the variable GDP_US in the dynSeries object ts. Note that this works on an arbitrary number of variables and that regular expressions can be used (see previous commits).