I am struggling to interpret a interaction term.

A 1% point increase in military expenditure growth, leads to a X% point change in GDP growth.

what is a 1% point increase in military expenditure growth? i.e. is it a 1% point increase in the rate of increase of military growth. For example from 5% to 6%.


is it 5% -> 5.005%

and how does this relate to the actual variable military expenditure not the military expenditure growth?

My model is of the form: my model

whereby the deltalog(variable) = growth of (variable)

My thesis supervisor has told me that "a 1 percentage point increase in the growth of military expenditure as a % of GDP, is associated with a beta percentage point increase in the growth of GDP per Capita."

I am just trying to wrap my head around the meaning of this? Would it be an increase in the rate of increase of GDP?

Any help greatly appreciated


  • $\begingroup$ stats.stackexchange.com/questions/337847/… $\endgroup$ – oneloop Apr 1 '18 at 20:40
  • $\begingroup$ @oneloop thanks, but rather I am after the intuition, as a 1% change is different to 1% point - thats the thing im after.. $\endgroup$ – Stat-metrics Apr 1 '18 at 20:48
  • $\begingroup$ If you can provide information on the data set and the model you are using, this will help others prepare their responses. $\endgroup$ – Gregg H Apr 1 '18 at 21:01
  • $\begingroup$ @GreggH Sorry about that i shall edit my original message $\endgroup$ – Stat-metrics Apr 1 '18 at 21:39

"Percentage point" generally refers to a unit that is invariant with respect to the starting point.

So in your example, a one percentage point increase would be from 5% to 6%.

A one percent increase would be from 5% to 5.05%.

  • $\begingroup$ I see as such would a 1% point increase in GDP growth rate of 0.1835, thus be 1.1835? $\endgroup$ – Stat-metrics Apr 1 '18 at 21:38
  • $\begingroup$ If the GDP growth rate is in a percentage scale. In other words, from 0.1835% to 1.1835% is a one percentage point increase. If 0.1835 is something other than a percentage, "percentage point increase" wouldn't have any particular meaning. $\endgroup$ – Patrick Malone Apr 1 '18 at 21:45
  • $\begingroup$ ahhh i see, if the case were that 0.1835 was the raw rate after computing (new GDP - old GDP/ new GDP), could i multiply that by 100 to make 18.35% and then talk about a percentage point increase as 18.35% to 19.35%? $\endgroup$ – Stat-metrics Apr 1 '18 at 21:48
  • $\begingroup$ I don't know economics, but wouldn't it be (new - old)/ old ? But in either case, yes, if you take a measure that is a percentage, then a percentage point increase increments the number by one. $\endgroup$ – Patrick Malone Apr 1 '18 at 22:00
  • $\begingroup$ you are correct, my fault! and thank you very much! $\endgroup$ – Stat-metrics Apr 1 '18 at 22:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.