Interpreting impacts of ratio (bounded) explanatory variables

[Ann Ng]

Hello,
I run regression with probit and get the table of average marginal effects. I am wondering how can I interpret the quantitative impacts of ratio (bounded) explanatory variables. For example, I have an independent variable: share of income on housing (X1) which is between 0 and 1, and its APE is 0.07 . The dependent variable is the probability of investment.
=> An increase in X1 by 1 unit leads to an increase in the average probability of investment by 7 percentage points. However, I find it odd to say that X1 increase by 1 unit as its value is bounded between 0 and 1. Do you think it is acceptable to interpret the results as above or what is a better way to interpret the results pls?
Thank you very much!

[Joseph Coveney]

What's APE?

Also, how can someone's probability of making an investment (presumably electively drawn from the person's disposable income) increase by seven percentage points when the percentage of that person's total income devoted to rent or mortgage payments goes from zero to 100?

[Ann Ng]

HI Joseph. APE is average partial effects. Please ignore the business meaning of the result. I just want to know how to interpret the number rather than the economic/business meaning in the variable at this point. Also, my independent variable is not in percentage, it is the ratio between 0 and 1 in the regressions. Thank you very much!

[Nick Cox]

This just hinges on convention. You could make your proportions into percentages. Then the effect of a unit change is the effect of 1% or 0.01. Depends what you want to report.