libmep new release

[mihai....@gmail.com]

Hi,

I have released a new major version of the libmep.

This update include a huge number of changes.

Notably it includes the support for time-series predictions.

In the "main" folder of the library there is a new example which shows how to find the formula of the Fibonacci serie. Also, the example show how to predict the next 10 values of the serie.

Documentation for the library is not updated yet, but I plan to include more code in the examples in order to show how to use most of the functions.

Mihai

[mihai....@gmail.com]

Hi,

I just published a new example of using libmep for multi-variate time-series.

https://github.com/mepx/libmep/blob/master/main/libmep_test_time_series_multivariate.cpp

Data are taken from this article 

Mehra Y.P., Wage Growth and the Inflation Process: An Empirical Approach. In: Rao, B.B. (eds) Cointegration, pp 147-159. Palgrave Macmillan, London, 1994

and can be downloaded from the "data" folder of libmep (wage_growth.csv file):

https://github.com/mepx/libmep/tree/master/data

No validation data is used; just training data.

Mihai