I have built a linear mixed regression model with fund returns (measured in percentage ie. 0.01 denotes one percent) as the dependent variable. For the independent variables I have percentage level of debt, Log(fund size USD), real GDP growth and 10 Year Real Interest rate.
I have set up the data so percentage level of debt, real GDP growth and 10 Y Real Interest Rate are all measured in percentages like fund returns with 0.01 equalling 1%.
I have the following coefficients:
0.599 (Real GDP),
-0.0046 (10 Year Interest Rate),
0.323 (Debt),
-0.00752(logFundSize),
My intercept is -0.00878.
I am confused about interpreting these outputs. From reading similar posts I believe that one percentage point change (so 1%) in real GDP equates to 0.599% increase in fund returns. Whereas, a 1% increase in Debt increases fund returns by 0.32% and 1% increase in interest rates leads to a -0.0046% reduction in fund returns. Am I interpreting this correctly?
Additionally, as I have logged Fund Size which was originally in USD could someone please advise me on how this would be interpreted?
Best Answer
Given that your outcome/depended variable is a proportion between zero and one, it would be more appropriate to fit a Beta mixed effects model. The linear mixed model may give results/predictions outside the permissible (0, 1) interval.
A Beta mixed model can be, for instance, estimated with the GLMMadaptive package; for example, check here.