Is it a problem if all my independent and control variables are insignificant?

[Thomas Lehnsherr]

The overall model scores like this:
R-squared = 0.0061
Adj R-squared = -0.0129

So the model seems useless, right?

[Felix Bittmann]

Well, apparently your independent variables are not really able to "explain" your dependent variable. Are there any bivariate correlations?

[Thomas Lehnsherr]

I don't know much about the bivariate correlations. How may I check it?

(Thank you for your answer by the way.)

[Felix Bittmann]

As long as the variables are metric try correlate, if not spearman (for ordinal ones).

[Thomas Lehnsherr]

I tried correlate and this is what I found for two of the control variables:



| GDPper~l Inflat~u
-------------+------------------
GDPpercapi~l | 1.0000
Inflationc~u | -0.0927 1.0000

(GDP per capita growth and Inflation)

When I added more of my variables:

| GDPper~l Inflat~u MAvolu~r FDIIndex
-------------+------------------------------------
GDPpercapi~l | 1.0000
Inflationc~u | -0.1174 1.0000
MAvolumecr~r | -0.0331 -0.0425 1.0000
FDIIndex | -0.1430 -0.1169 -0.0413 1.0000


This is not a significant correlation, is it?

[Felix Bittmann]

It is up to you to judge the strength of your correlations. You need to know your data and previous research to do so. If you are just interested in statistical significance, you can type

pwcorr var1 var2, sig