Fama French type summary tables (without asdocx)

[Ralice Fang]

Hello. I want to create a summary table similar to Fama-French, dividing my sample into 5x5 groups based on two variables, resulting in a 5x5 table where the mean of a specified variable for each group is displayed. Lastly, I want to see if the difference between the highest and the lowest groups is statistically significant. Here is an example:

I've searched online and found that asdocx can accomplish this, but it requires payment. Therefore, I attempted to achieve this task using different code.

bys year: astile PPE5 = PPErate, nq(5)
bys year: astile INV5 = INV, nq(5)

label define PPElabel 1 "low" 2 "2" 3 "3" 4 "4" 5 "high"
label values PPE5 PPElabel
label var PPE5 "Net PPE-to-asset"

label define INVlabel 1 "low" 2 "2" 3 "3" 4 "4" 5 "high"
label values INV5 INVlabel
label var INV5 "INV: Capex-to-asset"

table (PPE5) (INV5), statistic(mean DTD) nototals

In my sample, my two grouping variables are PPE and INV, divided into five groups from low to high. The numbers displayed in the table are the mean of DTD for each group.

data is as follows:

* Example generated by -dataex-. For more info, type help dataex
clear
input double year long gvkey float(INV II) double DTD
2009 1004   .0209472 1 3.6878670995020997
2010 1004  .08319487 1 3.7574052347303484
2011 1004  .05354027 1  2.962848772685523
2012 1004 .017124746 1  3.470066405037888
2013 1004 .012401142 1  5.505813943762509
end

The issue is that it doesn't display the high-minus-low and its significance, which are the outermost columns and rows in the example table. I want to ask if there is a way to modify the code to add the missing parts.