Iason Kynigakis, and Ekaterini Panopoulou, "Does Model Complexity add
Value to Asset Allocation? Evidence from Machine Learning Forecasting
Models", Journal of Applied Econometrics, Vol. 37, No. 3, 2022,
pp. 603-639.

In our analysis we use monthly data focusing on the period from
January 1980 to December 2019. 

Note that data are provided only for the predictors, as the stock and
bond series are proprietary data obtained from CRSP. The commodity
data are acquired from the IMF Primary Commodity Price System database
(https://www.imf.org/en/Research/commodity-prices). Details on the
data are reported in the paper (please see Tables A1 to A3 in the
appendix, for stocks, bonds and commodities respectively).

The file data_predictors.csv contains the predictors used to generate
the return forecasts. The series GSCI and CP are retrieved from Eikon
Datastream and CRSP respectively, and they are not included in the
file. The first row contains the names of the series. Further details
on the candidate predictors and their sources can be found in Table A4
in the appendix.

This file is zipped in the file kp-data.zip. It is an ASCII file in
DOS format. Unix/Linux users should use "unzip -a".

Please address any questions to:

  Iason Kynigakis (iason.kynigakis [AT] ucd.ie)

or

  Ekaterini Panopoulou (a.panopoulou [AT] essex.ac.uk)