dseries issueshttps://git.dynare.org/Dynare/dseries/-/issues2021-07-09T10:07:12Zhttps://git.dynare.org/Dynare/dseries/-/issues/49Add time aggregation methods consistent with chained indexes2021-07-09T10:07:12ZStéphane Adjemianstepan@adjemian.euAdd time aggregation methods consistent with chained indexes2021-09-10https://git.dynare.org/Dynare/dseries/-/issues/55Make subsref robust to deeper levels of nesting2023-05-23T17:22:49ZJohannes PfeiferMake subsref robust to deeper levels of nestingExecuting
```
load data_series.mat
ts = dseries(data_series,'1999Q1');
o = x13(ts);
o.transform('function','log');
o.automdl('savelog','amd');
o.x11('save','(d10)');
% run
o.run();
```
stores results in a `dseries` object in `o.results....Executing
```
load data_series.mat
ts = dseries(data_series,'1999Q1');
o = x13(ts);
o.transform('function','log');
o.automdl('savelog','amd');
o.x11('save','(d10)');
% run
o.run();
```
stores results in a `dseries` object in `o.results.d10`. However, trying to read out the subfield `o.results.d10.data` does not work as expected. Instead of a double, it will still return a `dseries`. The problem seems to derive from `subsref` only expecting two levels of nesting. This, by the way, also creates an infinite recursion problem in the Matlab variable editor, as shown in the screenshot.
![image](/uploads/da523510a35c57b65dddacdb36f50494/image.png)
[data_series.mat](/uploads/25ee186c37cf504f9c0b515bff8b5881/data_series.mat)https://git.dynare.org/Dynare/dseries/-/issues/52Decide whether to support unicode files2022-07-26T15:59:21ZJohannes PfeiferDecide whether to support unicode filesThe unicode file [GMM_data.csv](/uploads/8011ad7b1edbbb4996f72ba452b50231/GMM_data.csv) does not work due to a byte order mark (BOM) at the beginning. We use `importdata`, which assumes we are loading an ASCII file:
```
temp=importdata('...The unicode file [GMM_data.csv](/uploads/8011ad7b1edbbb4996f72ba452b50231/GMM_data.csv) does not work due to a byte order mark (BOM) at the beginning. We use `importdata`, which assumes we are loading an ASCII file:
```
temp=importdata('GMM_data.csv', ',');
double(temp.textdata{1}(1))
```
In contrast, `readtable` correctly detects unicode and drops the BOM:
```
temp2=readtable('GMM_data.csv');
double(temp2.Properties.VariableNames{1}(1))
```https://git.dynare.org/Dynare/dseries/-/issues/51Allow specifying xls_sheet2023-09-27T15:16:54ZJohannes PfeiferAllow specifying xls_sheetCurrently, `load_data.m` contains
```
if isglobalinbase('options_')
% Check that the object is instantiated within a dynare session so that options_ global structure exists.
% Should provide latter a mechanism to pass...Currently, `load_data.m` contains
```
if isglobalinbase('options_')
% Check that the object is instantiated within a dynare session so that options_ global structure exists.
% Should provide latter a mechanism to pass range and sheet to dseries constructor...
range = evalin('base','options_.xls_range');
sheet = evalin('base','options_.xls_sheet');
else
% By default only the (whole) first sheet is loaded.
range = [];
sheet = [];
end
```
This does not work with `makedataset` in `mom.run`, because `xls_sheet` is a subfield of `options_mom_`, which is not even a global variable.7.xStéphane Adjemianstepan@adjemian.euStéphane Adjemianstepan@adjemian.euhttps://git.dynare.org/Dynare/dseries/-/issues/50Change behaviour of eq and ne methods2023-12-14T20:02:25ZStéphane Adjemianstepan@adjemian.euChange behaviour of eq and ne methodsIn `dseries` class, to make them consistent with comparison methods introduced in 507ec822 (`lt`, `le`, `gt` and `ge`).In `dseries` class, to make them consistent with comparison methods introduced in 507ec822 (`lt`, `le`, `gt` and `ge`).Stéphane Adjemianstepan@adjemian.euStéphane Adjemianstepan@adjemian.euhttps://git.dynare.org/Dynare/dseries/-/issues/48Add interpolation methods to convert dseries objects to lower frequencies2021-07-15T09:15:23ZStéphane Adjemianstepan@adjemian.euAdd interpolation methods to convert dseries objects to lower frequencieshttps://git.dynare.org/Dynare/dseries/-/issues/42Extend dseries constructor for Excel files2021-11-11T13:21:39ZMichelJuillardExtend dseries constructor for Excel filesUsers want to load data from an Excle file by specifying sheet name and rang. I suggest to introduce the following syntax:
```
dseries(FILENAME, OPTION_NAME_1, OPTION_VALUE_1, OPTION_NAME_2, OPTION_VALUE_2,...)
```
The new options would ...Users want to load data from an Excle file by specifying sheet name and rang. I suggest to introduce the following syntax:
```
dseries(FILENAME, OPTION_NAME_1, OPTION_VALUE_1, OPTION_NAME_2, OPTION_VALUE_2,...)
```
The new options would be ``xls_sheet`` and ``xls_range`` as for Dynare ``estimation`` command
It would be good to also introduce option names for the existing options: ``initial_dates``, ``range_of_dates``, ``list_of_names`` (or, better, ``variable_names``), ``tex_names`` (or better ``variable_tex_names``)
and permit the syntax
```
dseries(DATA_MATRIX, OPTION_NAME_1, OPTION_VALUE_1, OPTION_NAME_2, OPTION_VALUE_2,...)
```
Note that we can distinguish between the existing constructors and the new proposed ones as the second argument is a ``date`` in the existing constructor and a ``char array`` or ``string`` in the new one