Reverse coefficients after regression on ihs transformed netwealth

[Lydia-Sophie Dufner]

Dear Stata Forum,

I am doing regressions with dependent var netwealth transformed with the hyperbolic sine. Literature says that you can transform these regression coefficients back into Euro value by using the following equation:

sinh(x) = {exp(x) - exp(-x)}/2

This is a trigonometric function that stata also already gives me (which I found out through this forum).

Friedline et al (2015) tells me: "This conversion facilitates drawing clear interpretations and practical implications from an otherwise complicated transformation. To accomplish
this conversion, the mean IHS value can be inputted into the following equation: 1/2 (e^1y+e^-1y)ßx" "

Still I am unable to transform a certain coefficient from my regression on ihs-transformed netwealth. I have no idea how to realize this with this information on sinh(x).
Does anybody have a useful idea how to actually work with this?

Thank you
Sophie

[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.

As you'd find it you looked for inverse hyperbolic functions:

asinh(x)
Description: the inverse hyperbolic sine of x
asinh(x) = ln(x +px2 + 1)
Domain: 􀀀8.9e+307 to 8.9e+307
Range: 􀀀709.77 to 709.77
161

With this, you should be able to get x and then do the equation.