Hi All,
I would like to do a hypothesis test on whether the average of weighted means of two different samples is the same. I have a data set that looks like this:
I compute the weighted mean of Leverage by country, using Sales as weights. I then take the simple average of all the years. I end up with something like this:
I repeat this same process for a different time period (2009-2014) -- but not the same companies or sample size! -- , and I want to test whether the resulting average values of each country in 2009-2014 are different from their 1997-2000 counterparts, and with what level of significance. I cannot figure out how to do this using ttest commands and would appreciate any help.
Thank you in advance,
Gonzalo
I would like to do a hypothesis test on whether the average of weighted means of two different samples is the same. I have a data set that looks like this:
| Company | Leverage97 | Sales97 | Leverage98 | Sales98 | Leverage99 | Sales99 | Leverage00 | Sales00 |
| Argentina1 | 0.47315782 |
6837541.66 |
0.87305553 |
8007910.66 |
0.6526006 |
3975080.83 |
0.36988563 |
6928492.83 |
| Argentina2 | 0.57015063 |
8152730.98 |
0.65441875 |
5917532.71 |
0.54195929 |
4242582.96 |
0.14777952 |
8440192.93 |
| … | ||||||||
| Brazil1 | 0.92833573 |
2300238.57 |
0.20469357 |
7750734.81 |
0.97714994 |
8431590.1 |
0.64225982 |
6880055.73 |
| Brazil2 | 0.20216352 |
5644301.99 |
0.43022949 |
1254098.8 |
0.02754164 |
9208091.99 |
0.93314635 |
9999704.17 |
| … | ||||||||
| Chile1 | 0.73741508 |
2805983.98 |
0.39869003 |
1472803.67 |
0.88297656 |
2717785.86 |
0.4897617 |
2137453.48 |
| Chile2 | 0.77459072 |
3948371.14 |
0.56826644 |
5621890.81 |
0.56917302 |
4749102.71 |
0.53418763 |
1666484.62 |
| … | ||||||||
I compute the weighted mean of Leverage by country, using Sales as weights. I then take the simple average of all the years. I end up with something like this:
| Country | Wt_mean97 | Wt_mean98 | Wt_mean99 | Wt_mean00 | Average (1997-2000) |
| Argentina | 0.56766019 |
0.93217148 |
0.14914068 |
0.89345066 |
0.635605752 |
| Brazil | 0.00028715 |
0.74773196 |
0.4243006 |
0.16871784 |
0.335259386 |
| Chile | 0.87908027 |
0.84415646 |
0.99564377 |
0.9176896 |
0.909142526 |
| … |
I repeat this same process for a different time period (2009-2014) -- but not the same companies or sample size! -- , and I want to test whether the resulting average values of each country in 2009-2014 are different from their 1997-2000 counterparts, and with what level of significance. I cannot figure out how to do this using ttest commands and would appreciate any help.
Thank you in advance,
Gonzalo