Hello fellow Stata users:
I have a panel dataset of patenting footprints of various companies. Each line in the dataset is a unique "company-year" identified by variables Company and Year. The other variables (columns) are patent classes and each value in the dataset is the number of patents that Firm X had in Year Y in patent class Z. So the patent data are arranged "horizontally" for each firm-year.
Now, here is my question. In Excel, I can easily compute correlations between any two firm-years. I simply type CORREL(..., ...) and then highlight the two horizontal arrays of firm-year patenting data. Is there a way to do the same in Stata? The "corr" command computes "vertical" correlations, that is correlations between specific variables across all observations. I need correlations between observations across all patent class variables. I have searched high and low but have not been able to figure out how to do it in Stata. Of course, I can export the data into Excel and compute those correlations manually in Excel but I would really like to be able to do it in Stata.
Any help will be greatly appreciated! Thank you! I will be happy to clarify if necessary.
Related Posts with Correlation between two observations across multiple variables
Changing lasso settingsDear fellow Stata users, I am running lasso regressions using such routines as lasso2 and cvlasso. …
p-value for standardized rate ratioHello, I would like to compare two directly standardized rates by computing a standardized rate rat…
Stacked bar graph per groupsDear all, I would like to realize a stacked graph bar showing the different values of each category…
another questions about converting HRF datetime to SIFI have read "help datetime", but am still having trouble converting a string datetime variable into …
xtpcse regression with fixed effectsHello, I have a panel of 50 countries over 25 years and I would like to perform a regression with pa…
Subscribe to:
Post Comments (Atom)
0 Response to Correlation between two observations across multiple variables
Post a Comment