Population Variable Signifcance

[JJ WM]

I am currently analysing a Log-Lin model, in which I am adding the independent variable LN Pop to test the significance within the regression model.
The regression output within stata states the p value > 0.05 significance level, therefore i do not reject the Null hypothesis, implying LN Pop has significance.
However the stata output for the lincom command regarding LN DPI and LN POP where DPI is consumers real income = total personal expenditure + real savings) provides a p value of 0.000.

The Null hypothesis i am testing is: Ho = LogPop + Beta y (income elasticity) = 1
The stata output states i should reject the Null, with Population having no significance, I am asking for help on how population can both be significant and insignificant and what is the implication of this? whether further regression is needed using per capita variations rather than aggregate.

Apologies for any confusion in the question, I am new and unfamiliar to stata and econometric regression, any help is appreciated.

Thanks,

J

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

You're testing totally different hypotheses - in the regression you test whether each parameter is different from zero, and in the lincom you test whether two parameters add to one. No reason you can't have any permutation of significance and insignificance with these tests.