In my master thesis, I want to examine the relation between CEO turnover and innovation. The first step is to identify 'why a CEO turnover takes place'. One of the variables are the consecutive loss years. Which means in words: After multiple consecutive loss years, it is more likely dat a CEO turnover takes place.
Now i have the data like this:
Company Name | Year | Net Income | Loss1 | Loss2 |
X | 2010 | 10 | 1 | 0 |
X | 2011 | 6 | 1 | 0 |
X | 2012 | 8 | 1 | 0 |
X | 2013 | 5 | 1 | 0 |
X | 2014 | -10 | 1 | 0 |
Y | 2012 | 14 | 0 | 1 |
Y | 2013 | 18 | 0 | 1 |
Y | 2014 | 22 | 0 | 1 |
Y | 2015 | 8 | 0 | 1 |
Y | 2016 | -4 | 0 | 1 |
Y | 2017 | -8 | 0 | 1 |
A company belongs in Loss 1 if the last available year has a negative Net Income, profits in all other years. So in the example above, Company X has a net income < 0 in 2014 and profits in 2010-2013, which means that Company X belongs to Loss 1.
Company Y has losses in the two most recent years available, 2017 and 2016, and profits in the other years, which means 2 consecutive loss years --> LOSS2. Logically, if the company had losses for 2017 till 2012, it will belong to the group: LOSS6.
Now I also have this examples in the data:
Company Name | Year | Net Income | Loss... | Loss... |
Z | 2010 | 4 | 0 | 0 |
Z | 2011 | 2 | 0 | 0 |
Z | 2012 | -10 | 0 | 0 |
Z | 2013 | 1 | 0 | 0 |
W | 2010 | 4 | 0 | 0 |
W | 2011 | -1 | 0 | 0 |
W | 2012 | 3 | 0 | 0 |
W | 2013 | -12 | 0 |
Loss 1: a loss in the most recent year of observation, profits in the other years (for company W means a loss in 2013, profits in 2010-2012, but it also has a loss in 2011 which means it does not follow the definition so it would not be grouped into a loss group).
Loss2: a loss in the two most recent years of observation, profits in the other years.
The data in my dataset is at most from 2010-2019, with some companies only having data from 2012-2019, some companies have data from 2010-2018 and so on.
I'm wondering if it is easier to only keep the data in a certain period. So lets say: Keeping only the companies with available data in the year 2010-2016.
Then I'm still struggling how to achieve the above situation.
I really hope you can help me out with this, since i've been working on it a lot but don't came to a solution. Hopefully you can!
Thanks in advance,
Remy
0 Response to Consecutive loss years in STATA
Post a Comment