How to Format Input Data for BankerΒΆ

Banker handles most conventional G7 data and matdat input formats, which are assumed to be known to the user. There are detailed discussion in G7 Help files and in the Appendix of Clopper Almon’s The Craft of Economic Modeling.

Banker also accepts the one-series-per-line format, which initially was developed to process very large data banks more efficiently. The general idea of this format is to stock all of the information about a series on one line (record), including, of course, all of its observations.

Here is an example of the one-series-per-line format:

gdp$  1985  Q  1  0  5025  5130.5  5.255e3  ...


1985:baseyear (can also be written as 85 [deprecated])
Q:frequency (M = Monthly, Q = Quarterly, A = Annual)
0:decimal point left-shift factor
5025:observation #1
5130.5:observation #2
5.255e3:observation #3

All seems clear except, perhaps, the fifth item under this format – the decimal point left-shift factor. Thus some explanation (or justification for it) is in order. Simply put, this left-shift factor enables you to manipulate (add) the decimal places in your data. For example, BLS data sometimes requires some additional decimal places, and this left-shift factor can help you to do just that. If, on the other hand, there is no need to modify decimal places in your data, then use zero as the left-shift factor.

Please take note that each item must be separated by at least one space (as shown in the example above).

It should be mentioned in passing that Banker will keep intact the series names exactly as in the formatted input file. That is, if a series name (or part of it) is in upper (lower) case, it will remain that way in the bank created by Banker.

Lastly, Banker allows G7-style commenting.


# This is a one line comment.
: This comment, however, is longer.
Clearly, it is a much more important comment,
For it takes three lines.
: Or four.

Previous topic


Next topic

Notes for Programmers

This Page