1: \begin{abstract}

2: \noindent {\rm This paper investigates the problem of selecting

3: variables in regression-type models for an "instrumental" setting.

4: Our study is motivated by empirically verifying the conditional

5: convergence hypothesis used in the economical literature concerning

6: the growth rate. To avoid unnecessary discussion about the choice

7: and the pertinence of instrumental variables, we embed the model in

8: a very high dimensional setting. We propose a selection procedure

9: with no optimization step called LOLA, for Learning Out of Leaders

10: with Adaptation. LOLA is an auto-driven algorithm with two

11: thresholding steps. The consistency of the procedure is proved under

12: sparsity conditions and simulations are conducted to illustrate the

13: practical good performances of LOLA.  The behavior of the algorithm

14: is studied when instrumental variables are artificially added

15: without a priori significant connection to the model. Using our

16: algorithm, we provide a solution for modeling the link between the

17: growth rate and the initial level of the gross domestic product and

18: empirically prove the convergence

19: hypothesis. \\

20: 

21: \noindent \emph{Index Terms} --- Variable Selection, Instrumental

22: Variables, Linear Model,

23: High Dimension, Sparsity. AMS: 62G08 \\

24: }

25: \end{abstract}