Traditional sectors with lower Ateco 1991 code, i.e. Food and Beverages, Textiles, Clothes,
Tobacco, tend to be underrepresented with respect to the original Capitalia sample, as we can see
from Figure A2. In any case if we consider High-Tech versus the others, there is an independent
distribution of frequencies in and out of sample (Pearson chi-square(1) test = 0.3952 with p-
value=0.530).
We then run a two-sample t test with equal variances to test for equality of average firm age
between the two groups (in-sample, out-sample). The results highlight that firms outside the sample
are on average 3 years older, and the difference in means is statistically significant.
Figure A 2. Distribution of firms by Ateco 1991 classification, in and out-sample
Distribution of firms by Sectors in and out of sample

out of sample I I in-sample
Group |
Obs |
Mean |
Std. Err. |
Std. Dev. |
[95% Conf. Interval] | |
Out sample |
-570 |
31.87 |
.938 |
22.41 |
30.03 |
33.72 |
In sample |
3469 |
28.87 |
.325 |
19.16 |
28.24 |
29.51 |
Combined |
4039 |
29.29 |
.309 |
19.67 |
28.69 |
29.90 |
“diff |
3.00 |
.887 |
1.259 |
4.742 |
Degrees of freedom: 4037
30
More intriguing information
1. Beyond Networks? A brief response to ‘Which networks matter in education governance?’2. Literary criticism as such can perhaps be called the art of rereading.
3. For Whom is MAI? A theoretical Perspective on Multilateral Agreements on Investments
4. Peer Reviewed, Open Access, Free
5. The name is absent
6. Design and investigation of scalable multicast recursive protocols for wired and wireless ad hoc networks
7. he Effect of Phosphorylation on the Electron Capture Dissociation of Peptide Ions
8. The name is absent
9. Herman Melville and the Problem of Evil
10. The name is absent